A preprocessing method for fruit and vegetable recognition based on visual subject detection

Abstract

The invention discloses a fruit and vegetable recognition preprocessing method based on visual subject detection. Aiming at the problem of difficult cropping and segmentation of fruit and vegetable images in complex environments due to factors such as uneven illumination and uncontrolled background, a method based on visual subject detection is proposed. preprocessing algorithm. Firstly, adjust image brightness and remove noise, then use manifold sorting to perform saliency detection to obtain a saliency image, and finally use gradient image and position weighting to fuse to obtain image cropping and complete image preprocessing. Through experiments, the fruit and vegetable images in complex environments can be correctly cropped, and better accuracy, robustness and real-time performance can be achieved.

Description

technical field [0001] The invention relates to the fields of food application and health, in particular to a fruit and vegetable recognition preprocessing method based on visual subject detection, which is used in the technical field of fruit recognition. Background technique [0002] With the continuous development of the field of artificial intelligence, people's lives are gradually becoming intelligent, and image recognition technology occupies an important position in the field of artificial intelligence. The invention studies the location technology of fruits and vegetables under complex environment and uncontrolled lighting conditions, uses the improved manifold sorting saliency detection algorithm, and integrates gradient energy and position information on the basis of the original technology to locate the fruits and vegetables. [0003] In the image processing stage, the original image cannot be directly applied to image recognition software due to factors such as a...

AI Technical Summary

Problems solved by technology

[0006] 1. Non-uniform scaling

Images