Anti Rob

Security is the main aspect that matters for any organization that is concerned with. In many of the organizations like banks, restaurants, offices, we can see the cameras that will be installed in order to record the footage within the organization. This will help at the time of any theft dete

Project Title

Anti Rob

Project Area of Specialization

Artificial Intelligence

Project Summary

Security is the main aspect that matters for any organization that is concerned with. In many of the organizations like banks, restaurants, offices, we can see the cameras that will be installed in order to record the footage within the organization. This will help at the time of any theft detection. It can act as evidence. There is an application that can help in detecting the thieves that are found through the Smart Cam footage just through their motion, it is possible through the use of Anti-Rob application. This will be one of the interesting applications that one can work on and implement in real time world

Here we propose a camera footage based theft detection along with thieves tracking based on motion. For this purpose, we use image processing to detect theft occurrence and motion tracking to track thieves in Smart Cam footage. The System allows enhancing theft detection without the use of sensors. This application uses foreground detector and blob analysis to detect motion of the thief. This system focuses on detecting objects. Here we don’t process the entire video, but we work on initial video, a frame in which the moving objects are segmented from the background. Here we applied image preprocessing steps in order to remove undesirable noise, and we used some image processing methodology to fill gaps in the detected objects. We used blob analysis methodology to detect connected component. The System rejects smaller objects. Finally, motion of large object is detected and surrounded by a bounding box. The System will detect the thief and beep sound will be generated, this can be later replaced by SMS notifications or remote alerting systems

Only the initial video frame will be used by this application. Then the largest moving object will be detected. A bounding box will be put around the detected object. The image processing method is used in this application. The cameras that are installed in the organizations will be continuously recorded and saving the footage. This application can help in providing an  enhanced version of security for the organization

generated bring into existence More (Definitions, Synonyms, Translation)

Project Objectives

• We are providing security for your surrounding for better monitoring and decrease the number of crimes.
• People are alerted by a beep sound if in case of any suspicious activity.
• Motion tracking for better and fast theft analysis
• Efficient use of image processing for the detection of thieve.
• This application is fully automated without any difficulty.
• The house owner can view the live stream of motion that takes place inside the house or any area by the stranger.
• A notification and alarm services are included to alert the owner about the strange actions.
• We are also using PIR sensor to detect the motion of objects.

Project Implementation Method

A Smart Cam framework comprises a few cameras, checking and recording frameworks and control room operations since Detection and Tracking of Objects and then Classifying it into Abandoned or not, is a process, which will require low level processing of an image frame. Image Frame is extracted at a certain frame speed (in FPS) from a video file. Each and every image frame undergoes the following steps:

Background model

It is not necessary that background remain the same at all the time, there might some changes occur due to addition or subtraction of some objects or there might be some changes due to change in light intensity or illumination, so we have taken the average of some background images at a regular time interval when no activity is going on, so that any new changes will get minimize. The resultant image is used as a background image for image Subtraction.

Low level processing

 Before proceeding with object detection and tracking, we have segmented objects of interest and minimizes the noise. For that we have done Image Subtraction to obtain a foreground image. Image Subtraction is done by subtracting Image with activity going on with the background model. There will be an introduction of some noise due to intensity difference or some minor change in positioning of objects.

Detection

In an image frame, there are both people and objects. So, to detect objects we have used two parameters: Area Based Threshold and InterPixel Distance.

Tracking and Classification

After detection of an object we have performed object tracking, in order to determine whether the object is suspicious or not. We have considered a threshold time, after which the object will be declared suspicious, and a beep sound will be generated.

We are using Wyze Smart Camera for detection of objects or any suspicious activity. It is also giving us night vision and 360 degree vision in under 3 seconds, the vertical range is 93 degree and horizontal range is 360 degrees.

We are also using one more camera for area surveillance and remote monitor. It gives you live stream to your phone. It is a security camera with motion and sound detection. This Wi-Fi camera detects movement and sound, records and sends notifications to you. Night vision and two-way audio: Automatically switches to night vision mode in low light when night falls. Night vision ranges up to 49 feet. Support two-way voice intercom, alarm push, PTZ control, mobile phone real-time video.

person a human being More (Definitions, Synonyms, Translation)

Benefits of the Project

Move Alerts 

The System will detect the thief and beep sound will be generated, this can be later replaced by SMS notification or remote alerting system This application can help in sending instant alerts related to the unusual activities taking place in the organizations with great ease.

Easy Access 

This application can be accessed anytime and anywhere from the world.

Automated 

This application is fully automated without any difficulty.

User Friendly 

This application will be user-friendly since the user interface will be simple and easy to understand even by the common man.

Motion Tracking 

This application can help in detecting the motion of the objects with great ease.

Voice Recognition 

In this module the risky voice analysis can take place. When the customer enters into the system the voice recognition is started. The API converts the voice into text. The Risky voice, strings are “Help” and “Emergency”.

Long Time Tracking 

The module takes input from the Human Detection. When the human enters into the system it calls the timer to measure the time. Start-Stop timer is the unique library used for time measurement. When the predefined time limit for the particular human detection is reached, the system sends the alert mail/message to the admin.

Camera Tempering 

In this module, the status of the camera is considered. If the lens is sprayed with glue and the image is darkened, it will not be possible to distinguish this from other situation where the same effect is seen. When this parameter is enabled, alert mail will be generated in all cases where the image turns dark or the lens is sprayed.

Human Collision 

In this module, the input is from the Human Detection. From the human detection module we know every human part is bounded by the respective bounding boxes and tracking is continued. If more persons are in the same environment, the multiple detection and tracking is also possible. So from that, the system has multiple boxes in the environment. If the collision occurs between the boundaries, then the violation may happen, so the system sends an alert message to the admin.

Technical Details of Final Deliverable

We are using Smart Cam footage for image processing, first image processing is done by an algorithm, and it detects the movement of any object or thief. If the camera observes unusual movements he will notify the user or admin by giving a notification or alarm. For alarm, we will use new equipment that will be any speaker or alarm and on the other hand, we will send a text message to the user as well.

The main process of detecting a human and an object uses the concept of background subtraction of the image with activity and an adaptive background model. The object detection is undertaken by the concept of Area based threshold and Inter-Pixel Distance, then the abandoned object tracking is done by setting a time threshold, once the object has been separated from the person for a time greater than the threshold time, it gets labeled as an abandoned object. Such a concept can be applied to other fields like locating lost baggage or lost cargo. There are many other possibilities for the further development of this algorithm based on application and system installation point.

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Security

Other Industries

Core Technology

Artificial Intelligence(AI)

Other Technologies

Sustainable Development Goals

Sustainable Cities and Communities

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com