Detecting Criminal Activities From CCTV by using Object Detection and machine Learning Algorithms

Now, a day’s Crime in every country is increasing day by day. Generally, Every day we listen to the news of different crimes of different categories like rape, assault, Kidnapping ,Robbery ,ATM Theft, Murders etc happening in different states, cities , countries. Almost all the newspapers, TV channels’, social media are filled with the news of Criminal activities happening all around the Whole World. In earlier times there is no method to detect Crime. After That the CCTV cameras were used to detect Crimes. But Watching these Videos manually by humans for detecting crimes is a very time Consuming process especially in today’s world of Artificial Intelligence and Machine learning .So this crime detection in CCTV surveillance becomes an important area of research in the field of machine learning. So, there is a very urgent need of the intelligent system which will detect the crimes from the real time CCTV Feed and classify them and provides an alert system to the nearest police stations and ambulances etc. So, that system will help in reducing the crime rate in any country. This paper reviews all prior research in this area, including approaches for object recognition and finding priority frames, techniques and algorithms like Yolo used to detect crimes , various datasets used and algorithms used to analyze crime data and train the dataset .It covers the various recent trends in researches in this field and analyzing the challenges faced and various research gaps and this paper also discuss how we can overcome these gaps in research so as to develop a better intelligence surveillance system in ml field.