Smart Person Tracking

Project Title

Smart Person Tracking

Project Area of Specialization Computer ScienceProject Summary

It is very difficult to the track a person which are known in the system firstly tracking the record of that person is a hard job then which route the person use for destination or whom he spends time most, the location of his most visiting place and detection of unknown person whom are not in system record.

In recent years, surveillance systems have gained increased importance in order to increase the safety and security of people. These systems have applications in various domains like home or bank security, traffic monitoring, defense; and in public places like railway stations, malls, airports, etc. Our goal is to develop an intelligent real-time surveillance system that can eliminate manual monitoring and help in increasing the efficiency of the system. In order to cover a large area, we need to install a greater number of cameras that leads to a greater number of videos that are to be monitored simultaneously. Therefore, it is of utmost importance to automate the complete system. In the proposed system, cameras are placed in such a way that there is a significant overlap between the field of view of the cameras. This helps in establishing an association between the cameras. The proposed real time surveillance system detects and tracks the person in motion and maintain its record and describe its mostly spending areas, routes and friends and time of entering and exit. The system provides different views in different cameras, thereby helping in resolving the occlusion in a simple and novel way. In this project we have worked on the problem of human detection, face detection, face recognition and tracking an individual. Our project is capable of detecting a human and its face in a given video and storing Local Binary Pattern Histogram (LBPH) features of the detected faces. LBPH features are the key points extracted from an image which is used to recognize and categorize images. Once a human is detected in video, we have tracked that person assigning him a label. We have used the stored LBPH features of individuals to recognize them in any other videos. After scanning through various videos our program gives output like- person labeled as subject1 is seen in video taken by camera1, subject1 is seen in video by camera2. In this way we have tracked an individual by recognizing him/her in the video taken by multiple cameras. Our whole work is based on the application of machine learning and image processing with the help of openCV, an open source computer vision library.                                    

Project Objectives

Smart Person Tracking system helps to monitor a given area of interest like some students play most the time as compare to study. Multiple cameras are used to cover a large area. In order to track objects successfully in multiple cameras, one needs to handshake among objects captured in multiple cameras. The key elements of our proposed surveillance system include change detection, track record of person , which route he use for destination, which people he spend time most, where he spend its time, detection of unknown person.

The goal of this project is basically to try to controls crimes and tracking person in university or any public area. Our project will mainly focus on the following objectives:

1. Video cameras tracking person by facial recognition
2. System that automatically track person wherever they go and maintains database
3. Track record of that person.
4. Which route he use for destination.
5. Which people he spends time most.
6. Where he spends its time.
7. Detection of unknown person

To achieve the first goal, we will be using OpenCV   library and python. For recognition of person datamining algorithm will use and system will use machine learning to adaptation of a person recognition. To achieve the second goal, This will help us to recognize the face of a person and information about the person will be save the time, place of a person will be saved in database.

Project Implementation Method

The input is given in the form of a video or a sequence of videos. The main objective of our project is to detect and recognize person and track them in a given video sequence. We have considered a few constraints in the interest of providing a successful output. Firstly, our work is mostly restricted to homogeneous domain i.e. the videos provided by the user must be shot with a homogeneous background i.e. plain background. Secondly, our system is successful in detecting humans in majority of the cases but it recognizes only those individuals for which it is already trained. So if there is need of recognizing new individual for which machine is not trained, machine must be trained for that individual using our program.

There are some assumption taken by us. We have assumed that in most of the cases the person
is facing the camera so that his/her face is successfully detected and recognized by our program.

There are certain major challenges that we faced while doing this project. Firstly, detecting
humans in all possible postures like- standing, sitting, lying etc was a great problem. Secondly, detecting humans in all possible orientations i.e. when their front side is facing the camera, when their back side is facing the camera, when they are facing the cameras sideways etc. Reducing the false positive rate to minimum possible value is still a greater challenge.

Benefits of the Project

One key advantage is that it does not require the cooperation of the test subject to work. Properly designed systems installed in airports, multiplexes, and other public places can identify individuals among the crowd, without passers-by even being aware of the system. Other biometrics like fingerprints, iris scans, and speech recognition cannot perform this kind of mass identification. However, questions have been raised on the effectiveness of facial recognition software in cases of railway and airport security.

In the 2000 presidential election, the Mexican government employed facial recognition software to prevent voter fraud. Some individuals had been registering to vote under several different names, in an attempt to place multiple votes. By comparing new facial images to those already in the voter database, authorities were able to reduce duplicate registrations. Similar technologies are being used in the United States to prevent people from obtaining fake identification cards and driver’s licenses. There are also a number of potential uses for facial recognition that are currently being developed. For example, the technology could be used as a security measure at ATMs. Instead of using a bank card or personal identification number, the ATM would capture an image of the customer's face, and compare it to the account holder's photo in the bank database to confirm the customer's identity. Facial recognition systems are used to unlock software on mobile devices. An independently developed Android Marketplace app called Visidon Applock makes use of the phone's built-in camera to take a picture of the user. Facial recognition is used to ensure only this person can use certain apps which they choose to secure.

Technical Details of Final Deliverable

For the first phase of human detection, the program takes a video or a sequence of videos as input. In the next step,  detecting objects with the help of openCV library to detect humans. As the program then detects humans and a rectangle is drawn around the human. In some cases our program draws more than one rectangle for single person because of false positive cases, to minimize such false positive cases.

Now, the program moves on to the second phase i.e. Face detection. For the detection of
faces the program loads the Haar features into the memory, which contains features of faces and using these features the program tries to detect human faces. After detection of human faces the program show detected faces in a separate window. Face detection is independent of human detection. There are certain cases in which human is not detected but face is detected, it happens because in some video only a part of human body is visible not complete body this phase gets completed here.

After the face detection phase, comes the phase of face recognition. But for this phase there is a prerequisite, to recognize a person firstly, we have to train our machine for certain images of that person. Here training is done using Haar cascades and Local Binary Pattern Histograms (LBPH). Firstly Haar features are used to detect faces in video frames, now each detected face is treated as a data set to train our machine and LBPH features corresponding to each faces are saved in a file. In our case it is yaml file

Face recognition is done by importing the saved features of yaml file into the memory, with the help of LPBH face recognizer program tries to recognize faces using LBPH features. Once faces are recognized the person is given the label of recognized face. So each recognized person is assigned a label. The name or label of recognized person is given as output along with the confidence factor. Confidence factor is a parameter whose value is zero if the images in training set and and test case are same, higher the value of confidence factor lower the given image is similar to that person. We have set threshold value of confidence factor, above which we can say the given image is not similar to the labeled person. Thus, our program detects human, detect faces, recognize faces and track human by recognizing this individual has passed through this camera. So we have output like subject1 has passed through camera1 and he/she also passed camera2 and so on. In this way we have tracked person    with     help     of            multiple           cameras.

Final Deliverable of the Project Hardware SystemCore Industry SecurityOther Industries IT Core Technology OthersOther Technologies Big DataSustainable Development Goals Sustainable Cities and Communities, Peace and Justice Strong InstitutionsRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	70000
A4 tech 16mp Anti-glare pk-710g. and GPU	Equipment	1	70000	70000