Video anomaly detection method and device based on semi-supervised learning
A semi-supervised learning and anomaly detection technology, applied in the field of computer program products to enhance the recognition ability
Active Publication Date: 2020-11-10
深兰智能科技研究院(宁波)有限公司
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] However, the contribution of the abnormal score in the above scheme is mainly determined by some prominent local features, and in some videos, the neural network needs to understand the overall video to determine whether an abnormality has occurred, so only some highly distinguishable local features are considered. may not be accurate
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0029] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.
[0030] figure 1 It is a flowchart of a video anomaly detection method based on semi-supervised learning according to an embodiment of the present invention. Such as figure 1 As shown, the video anomaly detection method based on semi-supervised learning in the embodiment of the present invention may include the following steps:
[0031] S1, the video data is sequentially divided into u×v frames of video images, wherein every v frame of video adjacent to the s...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The present invention provides a video anomaly detection method and device based on semi-supervised learning, which divides video data into u×v frame video images in sequence; performs feature extraction on each packet respectively to obtain corresponding video features; obtains according to video features The average vector and importance vector of video features, and according to the average vector of video features to obtain the mask of filtering highly distinguishable features, and according to the average vector, mask and importance vector of video features to obtain the dropout layer of neural network; according to The dropout layer and the video feature vector obtain the modified features, and obtain the training parameters according to the modified features; during the test, the modified features are obtained according to the adjacent packages and input into the fully connected network, and the score of each package is calculated, and According to the score, it is judged whether there is an abnormality in the relevant position. The invention can hide the most distinguishing part in the video feature to capture the overall information, and can highlight the highly distinguishable information area to enhance the recognition ability of the neural network.
Description
technical field [0001] The present invention relates to the technical field of video detection, in particular to a video anomaly detection method based on semi-supervised learning, a video anomaly detection device based on semi-supervised learning, a computer device and a computer program product. Background technique [0002] In modern society, video surveillance technology has become the most important means of security monitoring. However, ordinary monitoring video processing methods require managers to check the monitoring screen. When the monitoring data is large, it is very easy for a dedicated person to observe fatigue, and it is easy to miss detection. Therefore, judging whether there is abnormality in the video and locating the abnormal part in the video have become an urgent need for monitoring and management. [0003] In the related technology, a part of the video is input into the C3D network (3D convolutional neural network) to obtain the video features of this...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V20/46G06V20/41G06N3/045G06F18/214
Inventor 陈海波张雷武
Owner 深兰智能科技研究院(宁波)有限公司