Unlock instant, AI-driven research and patent intelligence for your innovation.

Abnormality detection method based on image recognition

An anomaly detection and image recognition technology, applied in the field of image recognition, can solve the problems of high cost and large impact on the recognition effect, and achieve the effect of improving speed and accuracy, improving efficiency and accuracy, and reducing interference

Inactive Publication Date: 2017-07-14
李刚毅
View PDF5 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

For a long time, it has been widely used to manually identify abnormalities in the collected pictures and images, which is costly and the identification effect is greatly affected by human factors (such as experience, fatigue, etc.)

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Abnormality detection method based on image recognition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0043] Exemplary embodiments of the present invention will be described in detail below with reference to the accompanying drawings. The description of the exemplary embodiments is for the purpose of illustration only, and in no way limits the invention and its application or usage.

[0044] Since the cameras and video cameras used for shooting are usually fixed at a specific position and face the detected object (such as vehicles, equipment, pedestrians, etc.) at a certain angle, the content of the picture or image is relatively fixed. However, due to the individual differences in each detected instance, the exposure difference caused by the different time of taking pictures, there may be multiple checkpoints in a picture or image, and the abnormal parts usually occupy the picture or image. The influence of factors such as the small screen ratio has greatly increased the complexity of automatic recognition technology. In order to solve the above problems, the present inventi...

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

No PUM Login to View More

Abstract

An abnormality detection method based on image recognition is disclosed, which comprises the steps of normalizing a picture containing a detected target to obtain a grayscale image; carrying out image matting using a trained target recognition model to cut out a detected target image from the grayscale image; carrying out binary classification of the detected target image by a trained binary classification model to determine a confidence value of the target image; and if the confidence value of the detected target image is not higher than a preset abnormality threshold value, determining that the detected target image is an abnormal target. The method can effectively reduce the feature dimension contained in the picture without reducing the image feature information by converting the image containing the detected target into the grayscale image; and can effectively reduce the interference caused by the non-detection target image information by cutting out the detected target image from the grayscale image. According to the invention, the abnormality is detected based on the shape, the abnormal target can be automatically recognized, and the efficiency and accuracy of the abnormality detection are high.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to an abnormal detection method based on image recognition. Background technique [0002] The technical background related to the present invention will be described below, but these descriptions do not necessarily constitute the prior art of the present invention. [0003] With the wide application of camera and video surveillance systems in industry, there is an increasing demand for efficient and accurate classification of collected pictures and images, and for finding abnormal factors in pictures and images. For a long time, it has been common to manually identify abnormalities in collected pictures and images, which is costly and the identification effect is greatly affected by human factors (such as experience, fatigue, etc.). Therefore, an effective automatic processing method is needed to identify abnormal factors in pictures and images. Contents of the inventio...

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
no application Login to View More
IPC IPC(8): G06K9/32G06K9/46G06K9/62
CPCG06V10/25G06V10/56G06F18/24
Inventor 李刚毅于湄
Owner 李刚毅