Face and License Plate Recognition System
Face and License Plate Recognition System consists of localization, detection and recognition. The data for this case is obtained through any video dataset or from real-time environment. This system first extracts, foreground and background key frames from captured video data. Expected facial and li
2025-06-28 16:27:11 - Adil Khan
Face and License Plate Recognition System
Project Area of Specialization Artificial IntelligenceProject SummaryFace and License Plate Recognition System consists of localization, detection and recognition. The data for this case is obtained through any video dataset or from real-time environment. This system first extracts, foreground and background key frames from captured video data. Expected facial and license plate images data are compared with the images in the database. If no match is found with the data, then an information is generated for the concerned department. Since a parallel stream is used, one for facial and other for license plate recognition thus our proposed system outperformed other existing system. This system identifies the face and maps the processed information to the assigned name for facial recognition and simultaneously, for license plate it assigns owner’s name and vehicle model from the database. This whole project is installed on a reconfigurable hardware or unit, interfaced with the supported language or software.
Our aim is to eliminate the manual verification of face and license plate that are allowed in a building or office. This system having HD camera on the front gate capturing a video of the cars and extract frames of them. Further faces are detected by implementing Viola Jones method and license plated are detected by edge detection method from frames and then those images get separated from background and save into database. Faces and license plates in the database are then sent for recognition. Recognition is done through MLP-ANN using BP algorithm. All the faces and license plates extracted from the camera are mapped with the faces and license plates in the database and with the help of that result authorization is granted.
Project Implementation MethodFace & License Plates Recognition via Modular ANN is implemented in three stages. In first stage, at the entrance when the car arrives the best frame is captured through the real-time video. The extracted image is then sent to be preprocessed which will highlight fine details of the image and remove unnecessary distortion. In the second stage face and license plate are detected individually through ANN. Further the detected images and would be extracted from the background and are resized for computation purpose for feature extraction. Facial features are extracted through Viola-Jones method while for license plate character segmentation method is used. Now these images are converted into a single vector format as input for the ANN. The data for training is fed to the Multi-Layer Perceptron Artificial Neural Network (MLP-ANN) to update the weights and biased values to produce the desired results in terms of recognizing a person's face and obtaining a driver's license through back propagation of ANN, produce the desired results of recognizing face and license plate After the training process, the trained data values (learning rate, weights and biased values) are used for the testing process. The data of all the registered faces and license plates are stored in the database. The testing phase will begin when our system has been trained through lots of data and how to handle the real time environment, so it will just make a prediction that will be closest to the data presented at the input. In the third stage, face and license plate recognition is done by using ANN. Now, for decision making a union is made as both face and license plate were treated separately using ANN. Then, the system will decide whether the person entered with vehicle or alone and will treat the situation accordingly. If the face and license plate images are matched with the stored images in the database then authorization will be granted and is displayed in the GUI. The total process is done through python software on raspberry pi and a wireless camera.
Benefits of the Project- Helps find missing people and identify perpetrators.
- Protects businesses against robbery.
- Strengthens security measures in banks, airports etc.
- Number plate can be recognized and checked against the database almost instantaneously. From this, it takes as little as 48 hours to issue a penalty notice.
- Cars used in terrorism acts could be easily identified.
A Respberry Pi based systemusing python as the platform for the developed application.the application is equipped with digital image processing module as well as ML based learning module for effictive localization of face and license plate recognition.
Final Deliverable of the Project HW/SW integrated systemCore Industry SecurityOther Industries Others Core Technology Artificial Intelligence(AI)Other Technologies OthersSustainable Development Goals Sustainable Cities and CommunitiesRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 50100 | |||
| Raspberry Pi 3 B | Equipment | 1 | 15000 | 15000 |
| LCD Screen | Equipment | 1 | 10000 | 10000 |
| Keyboard and mouse | Equipment | 1 | 2000 | 2000 |
| Raspberry Pi Power Supply | Equipment | 1 | 2000 | 2000 |
| 32GB Memory Card | Equipment | 1 | 1500 | 1500 |
| VGA Cable | Equipment | 1 | 400 | 400 |
| HDMI to VGA Converter | Equipment | 1 | 400 | 400 |
| IP Network Bullet Camera | Equipment | 1 | 8800 | 8800 |
| Documentation | Miscellaneous | 1 | 10000 | 10000 |