Face Recognition Based Attendance System with Room Monitoring and Control
Being in an era where the technology changes constantly and innovation are evolved. Everyone wants feasible and rapid response instead to wait in queue. Keeping this in a mind, this idea is evoked to create such cost efficient device which is not only save time and money but also maintain the decoru
2025-06-28 16:32:32 - Adil Khan
Face Recognition Based Attendance System with Room Monitoring and Control
Project Area of Specialization Artificial IntelligenceProject Summary PROJECT SUMMARY:Being in an era where the technology changes constantly and innovation are evolved. Everyone wants feasible and rapid response instead to wait in queue. Keeping this in a mind, this idea is evoked to create such cost efficient device which is not only save time and money but also maintain the decorum of an organization. The aim of this project is to capture the video of the students in 1D, convert it into 3D frames, in preprocessing the background noises will be remove, then Feature Extraction of the Frames will be applied using GLCM Method and Feature Reduction of Extracted Features through PCA for best fit of texture which have to be correlate through the SVM to the database to ensure their presence or absence, mark attendance to the particular student to maintain the record. The Automated Classroom Attendance System helps in increasing the accuracy and speed ultimately achieve the high-precision real-time attendance to meet the need for automatic classroom evaluation. The controlling and monitoring through the presence can help in power savings which will be achieve through the presence of the person in the room.
Project Objectives PROJECT OBJECTVIES:The aim of this project is to capture the video of the students, convert it into
frames, and relate it with the database to ensure their presence or absence,
mark attendance to the particular student to maintain the record.
The objectives of the project are given below:
1. Detection of unique face image amidst the other natural components such as
walls, backgrounds etc.
2. Extraction of unique characteristic features of a face useful for face
recognition.
3. Reduction of weak textures and taking the best fit texture.
4. Detection of faces amongst other face characters such as beard, spectacles
etc.
4. Effective recognition of unique faces in a crowd (individual recognition in
crowd).
5. Automated update in the database without human intervention.
6. Controlling and monitoring the room environment depending on the people
present in the room.
Project Implementation Method Project Implementation:The Implementation of the project consist of following steps:
1. Conversion of Image: The 1D Image will be converted in to 1D vectors which will be further transformed into 3D vector which contains all the properties of 1D vectors. Which will help to recognize the person in 360 Degree.
2. Preprocessing: The second step is to remove background noise from all images and resize all images to fit a standard size.
3. Feature Extraction: The feature is defined as characteristics of an object. Feature extraction is a method used to reduce the dimensions of an image while extracting all useful data from it. The extracted features are placed in a feature vector which is used for classification. we have used GLCM to extract features from each image.
The extracted features are:
- Energy
- Entropy
- Homogeneity
- Contrast
- Correlation
- Cluster Prominence
- Cluster Shade
- Variance
- Autocorrelation
- Dissimilarity
- Maximum Probability
- Sum Average
- Sum Variance
- Sum Entropy
- Difference Variance
- Difference Entropy
- Information Measures of Correlation 1
- Information Measures of Correlation 2
- Inverse Difference
- Inverse Difference Normalized
4. Feature Reduction: Feature Reduction will be achieving through the Principal Component Analysis (PCA) which is a statistical procedure that uses an orthogonal transformation which converts a set of correlated variables to a set of uncorrelated variables. PCA is a most widely used tool in exploratory data analysis and in machine learning for predictive models. Moreover, PCA is an unsupervised statistical technique used to examine the interrelations among a set of variables. It is also known as a general factor analysis where regression determines a line of best fit texture.
6. Classification: SVM is a non-parametric classifier used for classification and regression. It builds a hyper plane or a set of hyper planes in an infinite or higher dimensional space in order to work. Generally, SVM is applied to classify data into one of two classes, however it can be modified to classify data into more than two classes using 1-vs-1 or 1-vs-all technique.
SVM provides good generalization capability along with reduction in computational complexity as well as removal of over fitting of data. However, the training time of SVM is large compared to other learning algorithms and optimal parameters are difficult to determine when there is non-linearly separable data.
Performance of classifiers can be found by following evaluation parameters:
accuracy = (TP+TN) / (TP+FP+TN+FN)
sensitivity = TP / (TP+FP)
specificity = TN / TN +FN
7. Controlling The Room: Room Controlling will be achieving through the decision which will be given by the numbers of 3D face detected in the room.
Following is example of pseudo code
If Detected Face is > 1{
Turn on the light.
Turn on the fan
}
Else {
Turn off the all appliances // for 0 Face (no presence)
}
Benefits of the Project Benefits of the Project:Manual Time Clock or Punch Clock – it was very commonly used in the 20th
century. Every time the student entered in the class they had to use the manual
attendance. Teacher had to mark manually in his/her register.
Automated Time Attendance Software – fully automated solution that allows
teacher/faculty to monitor the students their time of entrance and leaving. All the
data are collected and then send directly to timesheets which makes time
management much easier and faster.
Poor manual systems of time and attendance management can lead to many
problems such as:
- Inconsistency in data entry and generate errors
- System is fully dependent on skilled individuals
- Time consuming and costly to produce reports
- Entry of false information
- Lack of security
- Duplication of data entry
There is a reason to be considered before you automate time and attendance
management and these are those good processes must already be in working
condition. Putting a new automated system in most cases won’t successfully
implement where one hasn’t existed before. The most effective way to get a
positive result from new automated attendance management software is where a
good manual time and attendance management system already exists and new
system automates these. It is also financially benefited to implement automated
time and attendance management software that is too simple, easy to use and
understand and requires no amount of external manual work.
Facial recognition is an AI-based technology that recognizes human faces. Sounds
quite easy, isn’t it?
In fact, the technology considers a number of aspects for successful face
recognition: jawline length, the shape of the cheekbones, eye sockets depth, etc.
All these factors help the technology “remember” who is the owner of the face.
In the heart of facial recognition technology lies the ability of the machine to
“learn”. So if you want to implement the technology in your processes, make sure
you have plenty of data to “feed” the machine.
Now that we are clear on the facial recognition working process, let’s see its
benefits
- Enhanced security
- Faster processing
- Seamless integration
- Automation of identification
- Massive data storage
3D Face Recognition based Attendance System provides the ability to Capture the image through the camera in 1-Dimension which will be transformed into 3-Dimension the pre-processing of 3D vector will be applied to remove the background noise, Then GLCM (Gray Level Co-occurrence Matrix) Method Will be the next process which characterize the Texture of an image by calculating how often pairs of pixel with specific values and in a specified spatial relationship occur in an image and then extracting statistical measures from this matrix then move forward to the next step in which Feature Reduction will happen through PCA Method. PCA (Principal Component Analysis) is a statistical procedure that uses an orthogonal transformation which converts a set of correlated variables to a set of uncorrelated variables to determine the best fit Textures which will passes throw the SVM (Support Vector Machine) Algorithm which is to find a hyperplane in 3-Dimensional space that distinctly classifies the data points. SVM perform pattern recognition between two classes by finding a decision surface that has maximum distance to the closet points in to Database. The detected face will be passes through the Recognition Classifier which will lead to mark the attendance of an individual. The Controlling and monitoring side of the project will use the detected faces which gives the decision according to the number if Faces Present in room this Doesn’t Depend on recognition but entirely depends on Detection of a person (Known or Unknown) and send the signal to the controller to perform their Operation.
| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 80000 | |||
| High Resolution Camera | Equipment | 1 | 6000 | 6000 |
| Graphical Processing Unit (GPU) | Equipment | 1 | 40000 | 40000 |
| Mechanical Structure | Equipment | 1 | 8000 | 8000 |
| Controller (Raspberry pi 3 Model B+) | Equipment | 1 | 7000 | 7000 |
| Auxiliary Components (Cables, terminal, Switches…) | Equipment | 1 | 9000 | 9000 |
| Miscellaneous (documentation Reports….) | Miscellaneous | 1 | 10000 | 10000 |