Human Activity Detection System For surveillance Using Artificial Intelligence (AI)

Project Title

Human Activity Detection System For surveillance Using Artificial Intelligence (AI)

Project Area of Specialization Artificial IntelligenceProject Summary                                                             Human Activity Detection System                                                                                                                                   Better Secuirty Better World

   The aims of Human Activity Detection System is to identify the irregular and suspicious activity and help the authority to react immediately when such activities is detected by using Aritifical Intelligence. It will categories the data into three classes.

1.safe

2.suspicious

3.criminal.

The safe activity involves standing, walking, laying, sitting etc.

Suspicious activity involves fighting or irregular activities and criminal activity involves criminal activities as shown in figure1.  

The prototype module consist of two components such as hardware and software. The hardware component consist of smart machine with surveillance camera which will be used to detect the humans and analyze their activity by using software component such as tensorflow, keras and openCv etc. During activity analyze if any suspicious and criminal activity is detected the system will send the report directly on the server so the authority can react immediately.

Project Objectives The Objective is to design smart system that observe the human activities frames by frames through cctv camera to see if there is any criminal activity and fighting  detected it will notify to the nearest police station Project Implementation Method                                                                   Implementation Method

1.  Gathering the image datasets of walking ,sitting,jumping, fighting, and weapons "https://github.com/satiety/fight-detection-surv-dataset"

2. Train the model by using dataset ( We are using Openpose Algorithm).

        3.  Testing the model 

Benefits of the Project

• Smart city: In future drones can be implemented to monitor human activities on a bigger scale to cover areas. The captured video data is useful to prevent the threats before the crime actually happens. These videos also become a good forensic evidence to identify criminals after the occurrence of crime in smart city.     
• Public or Religious Events: The security on these events is critical since there’s a huge crowd in these type of events and security is unable to prevent violent activities but with the help of AI we can prevent that.
• For the safety of tourism: With the help of this project we can protect the tourists from any threats or bullies

Technical Details of Final Deliverable

• The model can classified the data into three catagories:

1. safe
2. suspicious
3. Criminal

• if suspicious and criminal activity is detected the model send the report to the authority through server

Final Deliverable of the Project HW/SW integrated systemCore Industry SecurityOther Industries IT Core Technology Artificial Intelligence(AI)Other Technologies Artificial Intelligence(AI), Internet of Things (IoT)Sustainable Development Goals Sustainable Cities and Communities, Life on Land, Peace and Justice Strong InstitutionsRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	26500
Nvida Grpahc card	Equipment	1	25000	25000
Camera	Miscellaneous	1	1500	1500