Method and device for judging abnormity based on vision

Abstract

The invention discloses a method and device for judging abnormity based on vision, being applied to an intelligent robot. The method comprises the steps: obtaining a scene image; inputting the scene image into a convolutional neural network based on spatial gold sub-tower pooling to obtain a preset dimension feature vector; and inputting the feature vector of the preset dimension into a pre-trained anomaly detector, and determining whether an abnormal condition occurs or not according to a detection result of the anomaly detector.

Description

technical field [0001] The present application relates to the technical fields of image processing and intelligent robots, in particular to a method and device for judging abnormalities based on vision. Background technique [0002] In the process of performing tasks, intelligent robots often need to have various types of interactions with the surrounding scenes. This often requires the intelligent robot to pre-judge whether the current scene conditions meet the conditions for performing follow-up actions, and whether it needs to stop the current task for exception handling, for example: by adjusting its position or posture to meet the conditions for performing follow-up actions. Abnormality judgment is the basis for intelligent robots to have preliminary "intelligence". If the abnormality judgment method cannot quickly and accurately feed back the judgment results, it will cause the robot to run empty, which is manifested in the robot's interactive actions are blunt and dul...

AI Technical Summary

Problems solved by technology

However, this method requires the input of a fixed-size image, and the input image needs to be cropped or distorted. The cropped image may not contain the complete detection object, and the distorted image may be geometrically distorted, which will damage the accuracy of the judgment result.

In addition, the above method requires the detected object to occupy a relatively large proportion of the entire image, which limits the application scenarios of this method

Especially when it is necessary to use a small detected object as an abnormal and / or normal judgment standard in a large scene, because the detected object that provides key information accounts for a small proportion of the entire image , which often leads to the failure of the classification algorithm, and finally the correct judgment result cannot be obtained

Images