A system and method for classification of moving object during video surveillance
A technology for moving objects and videos, applied in character and pattern recognition, instruments, computing, etc., can solve problems such as low computing cost
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
example 1
[0088] Example 1: Human (walking)
[0089] Such as Figure 4 As shown, a system for classifying moving objects during video-based surveillance is tested. Allows a person to walk across the working area of video surveillance. Raw frames with the background are captured via a video capture device. Objects are segmented and extracted from captured frames. The extracted object is divided into two parts, ie, an upper half and a lower half. The center of gravity (C.G.) is calculated considering the upper half of the image, and the vertical line passing through C.G. divides the lower half into two parts, namely, the lower left half and the lower right half. Considering the lower left half of the extracted object, the variance of that part of the object residing in the lower left half from the vertical line passing through C.G. is calculated. Similarly, 16 consecutive frames were analyzed for calculating the variance of the center of gravity. In this particular example, the ...
example 2
[0090] Example 2: A moving two-wheeler (vehicle)
[0091] Such as Figure 5 When analyzing a moving two-wheeler (vehicle) in the video surveillance range as shown, following the same procedure explained in the first example, 16 consecutive frames were analyzed. In this particular case, the C.G. variance was calculated to be 1.6100 for 16 consecutive frames.
example 3
[0092] Example 3: A moving car (vehicle)
[0093] Such as Image 6 When analyzing a moving car in the video surveillance range shown, following the same procedure explained in the first example, 16 consecutive frames are analyzed. In this particular case, the C.G. variance was calculated to be 0.2400 for 16 consecutive frames.
[0094] Advantages of the invention:
[0095]1) The present invention provides a system for classifying moving objects during video-based surveillance, where only the set of centroids calculated from the sequence of frames has to be stored for variance calculation. There is no need to store object images from frame sequences. This saves memory space in the video surveillance system.
[0096] 2) The present invention uses less complex logic for classifying moving objects during video surveillance.
[0097] 3) The calculation cost of the present invention is low.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 