Face attendance + face recognition with Python | Computer vision tutorial

hey my name is felipe and welcome to my channel in this video i am going to show you how to build this face attendance system which is based on face recognition so let me show you super quickly how this works i am going to register myself into this system so i am going to put myself handsome and smile i think that's a very good picture i don't know what you think i'm going to put my name felipe accept user was registered successfully and now we'll try to log in so i am going to smile again and i had to wait a couple of seconds and then i get a message saying welcome felipe so this is exactly the tool i am going to show you how to build in this video welcome to my channel welcome to this video and let's get started so this is the system we are going to be building in this tutorial you can see that this is a very simple user interface these are all different windows we are going to be coding today this is going to be our face attendance system and you can see that this is basically our main window this is going to be the main window of our program and you can see we have two different use cases for this system we have two different buttons covering the two different use cases one of them is for users which are already registered into the system and those users could log in those users could press the login button and then for users which are not registered for new users and we also include a button to register a new user and then we have this huge label which is going to be your webcam so we are going to have the footage of our webcam in this window in this label and then we will have two different buttons covering the two different use cases of this system now let me show you what happens if we have a new user if we want to register a new user if a user clicks in the register new user button we go here to this second window and you can see this is very similar to the first window but now we don't have our webcam but we have the we have only a picture once the user clicks into register new user the system will take a selfie we take will take a capture of the webcam and it will put the picture here and then it's going to prompt the user to provide a name in order to to create a new registry in the in the database so the user will type something in the username and then click accept if he agrees with the capture and with the username and once the user clicks accept a new window pops up and it's going to say user was registered successfully and then accept but if the user is in this window and for whatever reason he doesn't like the picture that was taken maybe he didn't realize the system was going to take a selfie maybe he wants to he or she wants to prepare himself or herself in order to make this picture for whatever reason the user doesn't like this picture then the he can press try again and when the user press try again he goes to the main window so he can click register new user again so yeah so if maybe the user didn't know the system was going to take a picture and he he or she wants to prepare himself or herself in order to be a prettier or more handsome for this picture then he can just try again and the app will take a new a new picture a new capture in order to login into the database and it's very important because if we are talking about the face recognition software this is the picture we are going to match against our database this is very important this is the this is going to be the picture inner database and every time the user tries to log in into the system we are going to match it against this picture so this has to be a very clear and frontal picture of the user's face okay and then this is about the register new user window and we already covered this functionality but if the user is already is already registered into the system he will try to log in he will try to press this button and then he something will happen if that user is registered and we are taking a picture we are running or face recognition software and if the user is indeed in our database then we will welcome this user and we will create a log into the face attendance system we will say something like this user was logged in at this given time but what happens if the user is not in the database if some if an unknown user is trying to login into the system in that case we will prompt the user with a message like this saying hey who are you you are not logged into the system you are not registered into the system so please return a new user or try again because maybe he is indeed in the database but for some reason the face recognition software didn't work so he can try again pressing the login button again or he can just register our new user in case he is not already registered so this is going to be the app we are going to be building today you can see that this is a very very very simple using user interface all the different windows are very very very simple but it provides all the needed functionality we are going to need in this app in this software so this is going to be all for the design of the app we are going to be building today and something else i should say about this is that when you are about to build an user interface when you're about to build something like a software something else we are going to be doing today i strongly recommend you to do something like this i strongly recommend you to do like a system design something visual so you know exactly what you are building or what you are about to build because if you do something like this it's going to be super clear to set like the direction of where you are going with this software and it's also going to give you like uh like a visual assistance in order to know exactly exactly where you are at every given at every given time of your development process you know exactly what you need to do you know exactly what you are doing and you know exactly where you are going so doing something like this is going to be very very helpful and i strongly recommend you to do a system design even if it's something super simple like this for every project you work in okay so this is going to be the or system design now let me show you the face recognition software or the face recognition engine core we are going to be using in this tutorial this is a repository for face recognition and this is the face recognition machine learning software or i don't know how to say the most the face recognition engine we are going to be using um so i am going to post a link into this repository in in the description of this video in case you are interested in taking a look but this is going to be super easy to install we are just going to run a pip install which is around somewhere let me see if i find it install okay yeah a peep install and this library name and then for using this library it's going to be very simple too it's going to be enough to run a command like this or i think this is not even mandatory so we can just specify the two directories one of them with the people we know one of them is going to be your database and then the other one is going to be or unknown people so um this is pretty much the repository we are going to use and uh yeah and i'm going to show you in a few minutes how to use it and how it works so let's continue and let's go to pycharm because it's time to start working on our project and before installing the requirements and before starting code let me show you something first which is this util file i prefer this util file which contains a few functions which are going to be super super helpful for this tutorial this is basically for building the user interface this is only a few functions these are one two three four five functions and this is basically because we are going to be repeating the same process once and again for example you can see that we have to create many buttons we have many many buttons everywhere so it's going to be super super helpful if we have a function which is a get button so every time we call this function we are going to get a button and it's going to make everything super super easy or easier and yeah you're going to see what i mean in a few minutes and every time we call one of these functions we have to specify some parameters in the case of the get button we have to specify the text for the button right you can see each one of these buttons have a different text and also the color and also exactly what's the command what's the function we are running when we execute this button when we press this button so this is the utils file and this is very important we are going to be using it in our tutorial and then let's start with our today's tutorial let's start building this app so i am going to install the requirements this is a requirements file you can see it's only three libraries and one of them is the face recognition a library so i'm going to do bp install miners are requirements you can see i have already created a virtual environment this virtual environment is created with python 3.8 so that's already created and ready to be used and now everything is installed okay so i have installed opencv pilot which is another library for computer vision and face recognition and super quickly let me show you how this works you can see that from the repository the way to use this library sorry the way to use this engine is by specifying two directories let me see if i find it here and here it's the only two arguments we are putting into these functions so we are not specifying the tolerance we are not specifying anything we have to provide two directories one of them with the people we know one of them with our database and the other one with the people we don't know with the people we want too much to the people we know so i have prepared this directory which contains two folders one of them is called known people and the other one is called unknown people if i show you the known people you are going to see i have put four people in this directory one of them with a biden the other one is john snown from game of thrones the other one is this guy and then obama so this is only an example to show you how to use this app and this is going to be our database this is going to be all the people we know now let me show you what happens when i try to recognize this person this unknown person we know who this person is because it's like yeah we know who he is but this uh the the use case of this tool is that we are going to try to match unknown people to these people so i'm going to show you how this app works in order to match him in order to match this person this unknown person we don't know who he is into the people we know into the database so and the way this is going to work in order to give you like a very high level overview this face recognition software is going to try to match this face his face into each one of his of their faces and if he finds a match he's going to say oh yes i have found your match this is your person this is who he this person is and if he doesn't find a match it's going to say something like unknown person unknown user unknown uh whatever so let me show you i'm going to call face recognition this is how easy we can use it face recognition then the directory of for known people and then the directory of all or unknown people and i will have to wait a couple of minutes or a few seconds and that's going to be it so let's wait to see what happens and that's it so uh basically we are seeing here the path of four file we are seeing this path this is the the directory and this is the file name unknown person.gpg and then we have another word next to it which is obama and this word it this means that the this face recognition software has worked perfectly and it has identified what's the unknown person and if you take a look it has matched the the word we got from running this tool was the file name of this picture right so it has matched this face to this other face right i and and it had provided the file name for this picture let me show you what happens if i rename this as something random right if i do that and i run this again now we will have the same result but we are going to see a different value we are going to see all these string because this is the value of four new uh picture this is how we have renamed this picture so this is how this system works this is what you need to know in order how to know how this system works because this is how we are going to use it we are going to have our database which is all the people who has registered into the system it's not going to be with them this is only an example we are going to have uh other people for sure most likely uh and then every new picture we have into the system we are or every new user who tries to log in we are going to run this process and we are going to have the file name of the image in our database who matches the person who is trying to log in that's what we will try to do or that's what that's how this system works so we already covered the unit file we have installed requirements and i have show you how this face recognition software works so now let's start building our system i'm going to close this window and i'm going to close this one so i'm going to leave it open because we are going to need it for sure okay so the first thing i'm going to do is to import or library sdk okay and let's start by building uh let's start by building the main window so um and i think the best way to do it will be to define a class which is or app then i'm going to define the constructor which is going to be empty for now and then i need to define if name equal oh main this is a volume so it's double okay then we are going to create an instance of this class so it's going to be something like this okay and then we are going to run it and in order to run it it will be something like maybe for that it's going to be better to have another function something like start and i'm going to leave it empty and then it's just going to be upstart okay okay that's like the the most basic user interface or the most basic um everything is empty but that's going to be like the generic structure of our application okay now let's start creating our constructor we want to create a window and this is going to be achieved by calling um decay dot tk right we want to create a new window and it's going to be self dot main window okay so we have created our main window this is going to this is going to be a variable which is going to represent this window and now let's tell uh tinker let's tell python what's the size of this window so we are going to call geometry geometry and we are going to say the geometry for this window is going to be these numbers i have been doing many tests before starting this tutorial before recording this video and i have already all the values of the different sizes of the different windows and all the locations and so on so we are not going to lose time with that so yeah this is the size of our window is going to be 1200 times 520 and this is going to be the x and y location and this is basically this is going to be the size of our window of our empty window and now every time we want to store this application we are going to call a main loop okay so what we are doing here is just defining like this window the main window for now is completely empty and the other function we are defining is the start function and this is the one we are going to call in order to run our app so let's take this one step at a time and let's see how let's see what happens if i press play okay if i press play i get my window it's a completely empty window and that's all we have so everything seems to be working fine fine for now okay so yeah so now let's continue and what we need to do first is to create these two buttons because we have two buttons and this label so let's start creating our buttons let's say this is something like login button main window and it is going to be get button sorry util dot get button and this has import udl automatically and we need to specify the window and it's going to be self main window okay then the text the text is going to be login then the color if we follow the exactly same design i have specified in this uh design template is going to be green so we need to still this is going to be a green i think if we do something this is going to work yeah we can specify either text or we can just put the the rgb color if i don't if i'm not mistaken and then the command this is very important because every time we call we press this button we want to trigger a process we want to execute a function so i'm going to say let's say this function is going to be self.login and this it doesn't it doesn't exist for now but i am going to create it in a few minutes and that's basically uh all and then we have this default value which is foreground and this four one is basically the the color of our text but in this case we want to leave it in the default value because we want the text to be white so everything is going to remain like that okay so this is for or first button or login button and then i am going to create something similar but for the register new user button so we are going to have register new user button main window at button main window this is going to be register new user new user this is going to be grey remember this is grey and then every time we call this every time we press this button we are going to trigger register new user function that it doesn't exist for now but we are going to create in a few minutes and then in this case we are going to change the foreground we are going to change this default value and we're going to say this is going to be black because we want the text to be black in this case okay so everything is ready for the buttons and now what we need to do is to place is to locate these buttons exactly in this location for now we we haven't specified the location for these buttons and we will be able to do that by calling place and we will tell python to to place these buttons in x 750 and y 400 okay this is going to be the location no for a first photo no sorry this is going to be 300 and uh yes so this is where we are going to locate the login button and then for the register new user button we are going to locate it in 750 and 400. i was reading the other sentence a couple of seconds ago that's why i made a mistake and this is basically the these are going to be the location so for buttons and let's see if everything works properly for now it doesn't because i haven't specified these functions so let's do it super quickly by calling defining an empty function which is login and then another empty function which is register new user and this should work and now everything is like this is like the everything is starting to look like this we have a main window and we have two buttons and everything seems to be in exactly the location we want it to be so everything seems to be fine for now this is empty we haven't specified the login functionality or the register new user functionality so for now it's everything is empty but let's continue doing it this way let's focus on building all the interface all the user interface interface and let's code the functionality as the last step okay now i am going to create this webcam label and for this i'm going to create a new uh field which is cam label and for this i am going to call the get a mesh label function in our utils file i'm going to call this function and the only thing i need to specify is the window so i am calling this okay okay everything is okay and now we need to place it we need to call the same webcam label oh sorry dot place and in this case we are going to specify the x which is going to be 10 y it's going to be zero and in this case we need to specify the width and the height the width is going to be 708 is going to be 500 okay so this is pretty much it for the very basic functionality we don't really see anything in the label because it's an empty label for now but it's there trust me and what we need to do now is to put our webcam into the label into the label that it's we that we have just created and that for now it's empty so what i'm going to do is to call another function which is add webcam and i'm going to give it the label right this is going to be a function we haven't created yet but we are going to create it and we are going to specify this label why are we doing this because we need to add the webcam into the label so it's going to be very appropriate if we create a function for that so add webcam and this is going to be solve and label okay we input the label as parameter and in this function we are going to code everything we need in order to put the webcam into this label and for this is that we are going to call cb2 we are going to call it opencv because we need to call video capture which is the function in opencv that's going to allow us to access the webcam so i am going to access the full camera which is the number zero and i'm going to call this object dot cap right i'm going to call it cap and it's going to be a a class member so i'm going to call it like this but something important is that we are going to call this same uh function more than once or we may call the same function more than once so every time we call this function we cannot create a new object so what we are going to do is to ask if this variable is already created in the in the class member dictionary this basically is asking python if this variable was created and if it was not created we are going to create it and if it's already created then we are not going to do anything this is going to be very helpful and it's going to be a good ch sanity check now i am going to do things uh this way and i'm going to explain why in a couple of minutes i'm going to create another label another internal variable which is going to be called label this is going to be an internal variable we are going to use only as an auxiliary an auxiliary function but i'm going to call it this way and this is going to be equal to the label we are receiving as input and now i am going to call another function which is process webcam and this is the function we are going to use in order to put the webcam into the label the reason for making it this way is going to be clear in a couple of minutes so just stay with me and you're going to see why i'm doing it this way so i'm going to define this function process webcam and basically what we need to do here is to is to read frames from our webcam and then we need to put these frames into the label we need to do some some conversions some format conversions the images we are going to be reading they are going to be in the opencv format because we are taking these frames from video capture which is an opencv function and we need to convert all the frames all the images we are going to be reading from the webcam from opencv format into the pillow format which is the format we are going to need in order to put all the frames into the label we just created so i created a new function and that's what we are going to do in this function we are going to call self.cap dot read this is going to be empty and this is going to be red and frame okay so what we need to do first is to call this frame and what i'm going to do is this is the original frame we are getting and i'm going to call another variable which is going to be most recent capture and i'm going to call it like r i'm going to put an r at the end because this is going to be a numpy array and you're going to see why i'm defining this function like this in a couple of minutes and this is going to be my frame okay now i need to do a color conversion cv2 dot convert color i'm going to use my self most recent capture array and i am going to call color bgr to rgb okay so this is going to be only a color combination conversion this is going to be an auxiliary function this is going to be image and what i need to do then is to call now i need to start converting the image into pillow so i'm going to say something like from pillow from peel import image and also image decay which is the other object i am going to use and now i'm going to convert this into pillow so i'm going to say something like image dot image from array let's see what's the exact function name i usually work with opencv when it comes to image processing i'm not so familiar with pillow this is going to be image from all right yeah and we're going to input this new variable okay and let's say this is going to be the most recent capture but now this is going to be fill okay it's not really needed to define the two variables but it's going to be better to do it this way so we know exactly this is exactly the same information but now it's one of them is in the bgr format from opencv and the other one is in the pillow format in the format pilot feels comfortable with so let's continue and what we need to do now what now we need to continue uh converting this image and we need to create another object which is image decay and this is going to be equal to image decay which is going to be your image but in the tk format in order to process into in order to put it into our label this is photo image which is another object and we're going to call it image equal to this object okay this is the only part of this tutorial which i recognize it's a little like obscure because it's not really like there's not really a lot of intuition of what we are doing here we are just taking frames from the camera we are converting these frames into another format and then we are putting these frames into a label and we need to do things in a given way but it's a little arbitrary it's a little like it may be a little confusing so just bear with me and it's a few more instructions so we are going to call now this variable this internal variable we have created self dot label and we are going to call this member image decay okay and this is going to be equal to image tk to the object we have just created now i am going to call configure which is another member of this variable and i'm going to say this is equal to no i'm going to call this member with like this okay and then this is pretty much all we need in order to put one frame only one single frame into our label but we want to put the entire webcam we want to make these frames to look like a streaming we want to make it look like a webcam so what we're going to do and this is going to be the most important part of this function is we are going to call this function again after 20 milliseconds we are going to do this so this means that we are um calling this function again this the last instruction means that we are repeating the same process every 20 milliseconds so we are going to take a frame from the from the webcam we are going to read the frame we are going to convert this frame into the format we need in order to put it into the label we are going to put it into the label and once it's there we are going to repeat the same process every 20 min seconds so everything is going to look continuous everything is going to look like a video like a streaming and that's exactly what we want so let's see if this works properly i am going to open my webcam the number zero is my built-in webcam oh i made a mistake i only have to specify the name and that's yeah that's it so this is my webcam and this means everything is working properly now i'm going to do things uh the only change i'm going to do is instead of opening the number 0 i'm going to open the number 2 which which is an external camera i have logged into a system and that's like uh yeah that's better i like this camera the better because it has more it has a higher resolution so this is the camera i'm going to be using for this tutorial and that's pretty much all we need to do in order to create our main window everything is empty we don't have absolutely no functionality whatsoever these two functions are completely empty but i think we are it we are completely done when it comes to the uh with the aesthetics with the with the design with the with the user interface but everything is empty if i press here nothing is going to happen for now but everything is starting to look the way we want it so everything is going well and this is why i created this new variable and why i'm making this like this because we need to create a function with no input variables with no input arguments because we are going to call this function every 20 milliseconds and in order to call it or in order to call it like this we cannot specify any input argument into this function so the only way to do it or the way i find that i found to do it and it was a good solution was to make make it like this to define this or variable outside this function and then just calling this function from this other function so that's about it why i'm doing this this way with the webcam and we can just continue i'm going to start building or app i'm going to start adding our functionality and i'm going to start with the register new user i'm going to start with this functionality and every time we click on resistor user we should have a new window which is going to have a very similar appearance as our main window so let's start doing that i'm going back to this uh this is the function that's going to be executed every time a user clicks here and register new user and what we need to do is to open a new window so what we need to call is a function in order to that it's going to be very similar to this and the way we are going to do it is by defining a new variable which is going to be something like register new user window and this is going to be self main window dots sorry this is going to be tk dot top level and we need to specify this this is how we create a new window uh which is something like a secondary window right we have our main window and then we have this window which is uh inside the other window so and then we need to specify the geometry so i'm just going to copy and paste and actually it's a good idea to copy paste because it's going to be very similar but i'm just going to say 370 and 120 so what we are doing here is i'm creating a new window with exactly the same size but it's going to be exact slightly moved and shift to the right and to the bottom so doing this let's see if it works we click register new user and we have a new window but it's not really the size i want it so this is why i have copied and i i forgot to rename so now everything looks okay it's the same size as before but is slightly to the right and to the bottom okay and the only thing we need to do now is to start defining all these different fields so let's start with the buttons this is going to be very similar as before so i may just let me just copy and then i'm going to edit we have a new button this is going to be the accept button register new user window okay and this is going to be in this window the text will be something like accept it's going to be green and it's going to be accept register new user i need to define this function for now it's going to be empty okay and then the location of this new button will be let me take a look 750 300 exactly the same okay let's see if it works okay it works now i'm going to create a new bottom which is going to be a try again button and this is try again try again this is red and this is white yeah everything is okay and then this is 400 let's see what happens okay yeah everything is okay for now um okay so and now we have to grade my label and it's going to be similar to this other label i'm going to call it like this yeah so i'm going to go like this so this is going to be let's say capture label because now it's going to be a capture it's not going to be a webcam and this is going to be my new window and the location will be in 0 700 and 500 okay it's going to be exactly the same location as before okay and now as before the similar is going to be very similar as before we are going to create a new function which is going to be self add image to label right because now we want to add a single image we don't want to add a stream as before so this is what we are going to do and we are going to input this label okay now i'm going to define this function this is going to be like this it's going to receive a label and this is going to be very very similar as we did before so i'm just going to copy and paste some of the code what we need to do is to i'm going to change it better we need to now we need to work with one of these objects because we are already creating um remember we are taking captures all the time and we are updating the the main window all the time right we are um that's something like this remember this window we are taking captures all the time and we are just updating this uh picture this this webcam so what i'm going to do is take one of these captures and i'm going to take this one and i am going to create i'm going to copy paste these three sentences and most likely i will need to do something with them okay this is going to be my label um okay this should work let's see if it does okay it works if i yeah and the other one continues to work yeah everything is okay and something else i'm going to do is i am going to define a new variable which is going to be um the way i'm going to do it is self dots um register new user capture and this is going to be self most copy okay this is how i'm going to do it and i'm going to explain why in a couple of minutes i'm going to instead of adding this one i'm going to copy the one i get in the original format but it's going to be the same idea okay i'm going to explain why i'm doing this in a few minutes but uh for now let's just assume this is going to be necessary this is just another function another variable which is going to get the information for the same capture we are drawing into the window so this is all for now everything looks uh everything is empty now we are just defining this functionality but that's it now let's start uh coding the try again and they accept functionality let's find where is the try again i have never defined i never edited this field that's why we didn't have an error uh i was thinking why is it if i never defined the try again function uh it didn't come as an error when i when i executed this calls register news and that's why okay so today i'm going to return user and the only thing we want to do here is to call self dot register user window and call destroy why because we want to exit this secondary window and we want to get back to our main window so everything seems to be okay okay now they accept every time we accept sorry i forgot that we have additional fields which are the input username and also the entry field at the entry text so i'm going to code those i'm going to add those fields and let's start with the entry text so i'm going to define a new variable which is going to be entry text register new user and i am going to call util dot get entry text and that's why i created this function i'm going to input this variable and that's pretty much it for the entry text definition now i need to locate this entry text field and it's going to be located i need to call this dot place on this 750 and 150 okay so this is where the uh entry text is going to be located yes and now let's move to the let's see if everything works properly register new user yeah everything seems to be working properly now let's add the label the this text for this we are going to need a new label so i am going to define a text label register new user and this is going to be util get text label and we need to specify the text and the text is going to be please input username so underneath this okay and then um i need to locate this new variable and it's going to be located in the 750 and 70. so x equal 750 y equal 70. okay let's see if it works it doesn't work let's see why uh okay using the text label uh we need to specify also the window window and text and i was giving only the window only a text so i need to something like this every time we create an object using tinker we need to specify exactly in what window we are creating this object everything seems to be okay we are pretty much there we have made a lot of progress so far so what i'm going to do the first thing i'm going to do is to make it look nicer so i'm going to put the i'm going to put the input username in a different line so let's see if this works it's only a detail but i like it uh yeah let's just leave it like this or we can put the username in a new window uh in order to make it exactly like the requirements no i like it better before yeah let's just leave it like that okay so we have pretty much all and now let's accept every time we accept every time we press this button what we are going to do is to take this a picture and take whatever information the user has input here and we are going to update our database and or database remember is going to be something like this we need a folder a directory with or known people so we are going to have a directory with pictures and the name of the picture the file name of this picture is going to be how we want to specify this so this user in the database so we need a picture and we need a file name with the picture identifier that's what we need so what i'm going to do is go into the code directory and i'm going to create a new directory but i'm going to make it from here we need a directory with our database we need a directory where we can save our images so i am going to define a new variable which is going to be db directory and this is going to be located in my current directory db yes yes this is a good name for the db directory and if os dot path exists self.dvdr if not then i need to create it als make dear self dvd right if the directory doesn't exist i need to create it and if it does exist i can just leave it and i can just use the images that are inside that's something that needs to be specified what happens if when you execute this app when you execute this software if there are people who have already registered into the system do you want to use them though or do you want to start from scratch let's just use it and i'm not going to delete the directory if it's already there so let's just use it okay so this is going to be our directory or database with known people with known users so i need to go back to the accept let's see where i have the function here accept register new user now we need to specify the behavior now every time the user clicks accept and basically what we need to do we need to take the name the user has input into the entry field and in order to do that we are going to call uh let's see what's the variable name we need to call this variable and we need to call gate and this is very obscure 1.0 and then sorry [Music] 1.0 and then and 1c okay yeah that's the text we need to input i'm not sure why but that's how it works so calling this command is that we are going to get the information that the user has input into the text field so now let's continue and what i'm going to do is to define i'm going to save my image i'm going to save the image in the uh in the label the image that we have plot into the window and in order to do that i'm going to call cv2 in grid and where am i going to save this image i'm going to save it in dotpath.join the self.dvdr and the name is going to be formats [Music] formats name and this is going to be this dot jpg okay so this is where i'm going to save the image and what image i am what image am i going to save i am going to save the image i have created here let me find it register new user capture and this is why we have created a new variable because remember that the remember that our main window is capturing data all the time it's capturing is taking frames from from our webcam all the time continuously so we are taking one of these frames and we are putting this frame into our second window but we cannot use this one anymore because from the time this was plot from the time this was um from the time the system has taken this this frame it has taken many other frames many other frames continuously remember if we go back to the main window i'm doing stuff i'm moving and it's taking frames every single time so if we attempt if we try to use this variable it's not going to work because it's going to take the most recent capture which is not the capture we have put into this window and the user and we want to save this image because this is the image the user has validated as its selfie so what we need to do is to take this capture it's already in the opencv format so we will not need to do anything with it so let's see if this works so far register user please input username i'm going to say something like felipe accept and something didn't um work and that's because i mixed this up and it was the other way around okay only a detail oh and i need to release okay let's see registering user felipe accept i don't have any error let's go to my database and now i see a new file which is felipe it's this capture and it's uh yeah everything is working pretty pretty well so um if we go back to the um design to the system design when the user clicks accept not only we have to save the information not only we have to save the image and the username but we also need to prompt the user and to let the user know that everything has that everything has happened successfully that the user was registered successfully so after we save the image we need to call another function which is util.messagebox and we need to let the user know that um something like let's see if there's something user um register let's put the description first user was register successfully and then we need to specify the title for this message box and this will be something like success okay so yeah and this is basically why i have defined this new function this function in my youtube file and this is going to be super super helpful i guess i guess you realize why every time we need to plot one of these message boxes one of these windows one of these small windows which are basically a message to our user and we don't really need to do anything with them they only have like a default accept button so every time we need to to display one of these windows is that i am going to call this function in my utils file so i am going to do this and then immediately after i am going to destroy the the register new user window because we don't need it anymore i'm going to call self dot register user window dot destroy and let's see what happens let's see if everything works properly so registering user i prepare myself well this is uh i'm not sure if it's the same the the best capture ever but it's going to work felipe accept user was registered registered successfully okay and we go back to the main window okay so the user was registered successfully um this is 2s yeah okay so everything seems to be working fine we are almost there we have pretty much i would say 80 85 percent of all the process completely ready so everything is going uh great and we have completed the register new user functionality now let's go back to the webcam let's go back to the main window and now is that we are going to implement the face recognition feature the face recognition core and we need to specify what happens when the user clicks login so let's go back there now when the user clicks login we are going to have a directory which is the database directory which is going to have many many many pictures from all four users and we need to do some we need to compute all this logic for the face recognition and basically what we need to do is to call this same command we have called here that's basically what we need to do and let's do it this way because there are many ways to execute a command from python but as we need the output from this command we need to know exactly what image in our database was uh found us or was recognized so what we need to specify is we need to call the subprocess dot check output check output and we need to specify oh we need to specify it like this like a list of all the different words we need to offer command so every word every element in this list is going to be one of these words so the first one is face recognition which is our command and then it's going to be or images directory or database directory this is our directory with all all four known people and then we can specify the the the directory with all the unknown people or we can also specify the specific part of the image we want to we want to recognize so i'm going to put something like unknown image path okay so this is the command we need to to execute this is going to be output and now let's define this image this is going to be only temporary because the only thing we need to do is to save the image and then we need to we need the image into our hard drive in order to execute this command but once we have executed we don't need the image anymore so i'm going to define it and then after before leaving this function i'm going to remove it so it's going to be os dot remove remove yeah and i'm going to give it yeah that's going to be perfect and then in this location is that i need to call cb2 in right and what's the image i am going to write is the image again is the most recent capture but is this array right i'm taking selfies i'm taking images continuously and i'm going to take the most recent one in order to compare with my database so i am going to call self.most recent capture r and i am going to write this uh image let's see where am i where am i logging here so something like this okay we are pretty much there and now what i'm going to do is to print output because i want to see how output looks like what the output looks like for in all four different situations i'm going to see what the output looks like when the user is in fact in our database how the output looks like when the user is not in our database and when the there's going to be another situation which is when no face was detected in the face the user is trying to recognize so i'm just going to plot i'm just going to print outputs and i'm going to take a look what output looks like in all different situations i already got an image of me but i'm going to make make it look nicer so i'm just going to delete this image and i'm going to register it again but not try again so i am going to make everything looks i'm going to remove the mic for up for a second well i think it's like okay i'm not saying it's the best image ever and it's definitely not but it's going to work so this is going to be felipe accept user was registered successfully a new image was created and i'm going to try to log in and let's see what happens if i put myself into the camera the face recognition software should be able to detect me and if i go back to the terminal i see this is the output i get when the uh when the software detects me right it says this is felipe and um yeah so everything is okay now let's do it again and this time i am going to put an empty i'm going to cover my face so there should be no faces detected in this image let's see what happens no person's found this is the output we got if i cover my face so if no faces if there are no faces in the image we get a no person's found and now i will see what happens if i if if i'm not in the database so the way to do this is by deleting this image so i am not in the database anymore i will try to log in and i will see a message saying no unknown person because it has detected someone is has detected a face but this face is not in the database so these are the three different options we got and what we need to do now is to parse this string so we know exactly what we need to display to the user and what we need to do okay so the way we are going to do that is by splitting output into split into the comma because all these values they have a comma and i'm going to take the um second element okay i'm going to make sure this is a string the first thing i'm going to do i'm not completely sure okay and this is going to be output i'm just going to say this is going to be the name okay and this is going to be the first element and then i need to remove this so i'm not completely sure if this is going to be minus one or minus two or minus three i'm i'm going to put minus one and i'm going to see i'm going to print name again let's see what happens i will try to log in without being registered and let's see what's the name we get unknown person yeah i i for some reason is not taking this as a string so this is not a character but this is a i'm taking like all the individual characters anyway this it should be a three another one so we are reading all the values until the minus three i'm going to do it again and if i try to log in now i get uh unknown person and i'm not getting anything else if there's no face detected and i try to log in now i get uh no person's found perfect and if i register myself i will try again i will register myself and this is felipe accept user register successfully and now i will try to log in and i get felipe so everything is perfect let's continue so if name in one of these two values if there is an unknown person or if it's not persons found we will tell the user to try again so in this case we are going to call the util.message and we are going to tell something like oops and the message is unknown user please register new user or try again known user please register new user or try again okay and if the user is in our database else what we need to do here is to welcome this user so we are going to output another message and then we are going to tell the user something like welcome back something like that in the title and then in the description welcome um and it's going to be format on the name right it should work let's see if it works i'm already registered so i try to log in and i get welcome felipe okay and if no persons are found on user please return use or try again perfect yeah perfect everything is almost there we are almost there so this is i'm going to add a last character a point about and what we need to do in this situation too is to log the system into log the user into the system because remember this is a face attendance software so we need to log all the attendance of for people or for users and i'm going to create another member which is log path and this is going to be the path of for log we are going to use this file in order to log all of our users in order to log the data of all of our users so if this happens if if we have recognized successfully all or user i'm going to call with open soft dots log path this is going to be append because we are going to be logging people all the time so we cannot overwrite this um this file we need to append new information this is going to be as if and if dot right this is going to be we need to lock the name who is logging in and we also need to log the time okay and the new line character okay this is going to each entry in or log it's a very simple log it's not um it's necessary sophisticated we can definitely make it more sophisticated but for now i think it's going to be enough so this is going to be format name and then we are going to log the exact time this user is logging in so we are going to import date time something like uh yes we need to do it like that let's make it pep compliant and then this is going to be the datetime.datetime dot now okay the exact time the user is logging in and i think this is going to work let's see if it does login i'm not sure if it's going to recognize me he's recognized me this face recognition software works very well okay and then if i go back to the log i see felipe on the exact time felipe is logging in so everything works uh pretty well let's try to log in again let's see if we have a new entry reload and yeah a new entry so everything works perfectly and everything is ready everything is done so we have created this amazing app this very simple user interface but at the same time amazing app and in order to show you like more comprehensively how this works i am going to execute this um app this software this tool but i'm going to do it with more people the thing is i live alone so there's no one here to test this app with me and i don't have any friends so i cannot show you with real people but i'm going to test it with captain picard i guess you are familiar with capitan car he's the captain of the uss enterprise and i'm also going to test it with frodo so yeah these are the two people i'm going to test it with so i am going to i have already registered myself into a system if i go to leave it there's my picture and we already noticed this picture is well recognized so i'm going to do the same but now with captain picard and i'm going to put it like this i don't want to be on the picture i think there is going to be a okay register new user i'm going to try it again because i want a very clear picture i think that's going to be okay let's do it again uh like this right like this register user i think that's going to be okay this is going to be captain p chord uh accept user was registered successfully we have a new entry and now i'm going to do the same with frodo so this is frodo register user let's we again let's do it again i don't want any glare i will kill to me my speculator [Music] okay i think that's going to work so this is going to be frodo okay accept uh successfully and we should have a new entry with with fraud okay so now i will try to log in this is me this is like always i should be recognized as i was recognized before welcome felipe everything is okay now let's see what happens if frodo tries to login into the system i am going to something like this login and if i wait welcome frodo so yeah everything is okay frodo has logged into this face attendance software and now let's log captain picard uh something like this login and let's see what happens with capital welcome captain picard so uh and now if i go back to our logs i see i have this entry which is the the newest one then frodo and then captain picard so everything is working properly and this is going to be all for this tutorial this is how you can build a face recognition software and face attendance system using python and the face recognition library i show you in this video if you enjoyed this video i invite you to click the like button and i also invite you to subscribe to my channel my name is felipe i'm a computer vision developer and in this channel i make tutorials exactly like this one where i show you different applications of computer vision this is going to do it for this video and see you on the next video

Channel: Computer vision engineer

Views: 105,866

Keywords:

Id: z_dbnYHAQYg

Length: 75min 29sec (4529 seconds)

Published: Mon Sep 19 2022