{"id":78316,"date":"2020-06-02T11:04:26","date_gmt":"2020-06-02T05:34:26","guid":{"rendered":"https:\/\/data-flair.training\/blogs\/?p=78316"},"modified":"2021-08-25T13:54:57","modified_gmt":"2021-08-25T08:24:57","slug":"deep-surveillance-with-deep-learning-intelligent-video-surveillance-project","status":"publish","type":"post","link":"https:\/\/data-flair.training\/blogs\/deep-surveillance-with-deep-learning-intelligent-video-surveillance-project\/","title":{"rendered":"Deep Surveillance with Deep Learning – Intelligent Video Surveillance Project"},"content":{"rendered":"

Surveillance security is a very tedious and time-consuming job. In this tutorial, we will build a system to automate the task of analyzing video surveillance. We will analyze the video feed in real-time and identify any abnormal activities like violence or theft.<\/p>\n

There is a lot of research going on in the industry about video surveillance among them; the role of CCTV videos has overgrown. CCTV cameras are placed all over the places for surveillance and security.<\/p>\n

In the last decade, there have been advancements in deep learning algorithms for deep surveillance. These advancements have shown an essential trend in deep surveillance and promise a drastic efficiency gain. The typical applications of deep surveillance are theft identification, violence detection, and detection of the chances of explosion.<\/p>\n

Network architecture:<\/strong><\/p>\n

We have generally seen deep neural networks for computer vision, image classification, and object detection tasks. In this project, we have to extend deep neural networks to 3-dimensional for learning spatio-temporal features of the video feed.<\/p>\n

For this video surveillance project, we will introduce a spatio temporal<\/a> autoencoder, which is based on a 3D convolution network.\u00a0The encoder part extracts the spatial and temporal information, and then the decoder reconstructs the frames. The abnormal events are identified by computing the reconstruction loss using Euclidean distance between original and reconstructed batch.<\/p>\n

\"spatial<\/a><\/p>\n

Intelligent Video Surveillance with Deep Learning<\/h2>\n

\"deep<\/a><\/p>\n

We will use spatial temporal encoders to identify abnormal activities.<\/p>\n

The dataset for abnormal event detection in video surveillance:<\/h3>\n

Following are the comprehensive datasets that are used to train models for anomaly detection tasks.<\/p>\n

CUHK Avenue Dataset:<\/h4>\n

This dataset contains 16 training and 21 testing video clips. The video contains 30652 frames in total.<\/p>\n

The training videos contain video with normal situations. The testing videos contain videos with both standard and abnormal events.<\/p>\n

Dataset Download Link:<\/strong>\u00a0Avenue Dataset for Abnormal Event Detection<\/a><\/p>\n

UCSD pedestrian Dataset:<\/h4>\n

This dataset contains videos with pedestrians. It includes groups of people walking towards, away, and parallel to the camera. The abnormal event includes:<\/p>\n