Apple defect detection method and system based on convolutional neural network

Abstract

The invention provides an apple defect detection method and system based on a convolutional neural network, and belongs to the technical field of fruit defect detection, and the method comprises the following steps: obtaining image data of an apple; preprocessing the acquired image data; and inputting the preprocessed image data into a preset convolutional neural network model to obtain an apple defect detection result, wherein the preprocessing of the acquired image data comprises the following steps: removing an apple background by adopting a maximum between-cluster variance method, and adjusting the image resolution to be a preset pixel. The influence of pedicels and calyxes on defect detection accuracy is reduced, and efficient and accurate defect identification of apples can be realized.

Description

technical field [0001] The present disclosure relates to the technical field of fruit defect detection, in particular to a convolutional neural network-based apple defect detection method and system. Background technique [0002] The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art. [0003] During the growth of apples, various factors often cause defects on the surface of apples, which affect the appearance of apples, and even make them lose their edibility, which greatly affects the quality and sales of apples. It can be seen that the detection of surface defects on fresh apples is particularly important. Traditional detection methods are mostly manual, time-consuming, labor-intensive, and inefficient, and cannot meet the needs of mass production. Therefore, researchers began to seek to develop a fast, non-destructive and efficient apple surface defect detection method to achie...

Images