计算机工程与应用 ›› 2020, Vol. 56 ›› Issue (11): 10-16.DOI: 10.3778/j.issn.1002-8331.2001-0248

• 热点与综述 • 上一篇    下一篇

基于机器视觉的水果品质检测综述

何文斌,魏爱云,明五一,贾豪杰   

  1. 1.郑州轻工业大学 机电工程学院,郑州 450002
    2.广东华中科技大学工业技术研究院 广东省制造装备数字化重点实验室,广东 东莞 523808
  • 出版日期:2020-06-01 发布日期:2020-06-01

Survey of Fruit Quality Detection Based on Machine Vision

HE Wenbin, WEI Aiyun, MING Wuyi, JIA Haojie   

  1. 1.College of Mechanical and Electrical Engineering, Zhengzhou University of Light Industry, Zhengzhou 450002, China
    2.Guangdong Provincial Key Laboratory of Manufacturing Equipment Digitization, Guangdong Huazhong University of Science and Technology Industrial Technology Research Institute, Dongguan, Guangdong 523808, China
  • Online:2020-06-01 Published:2020-06-01

摘要:

随着经济的发展和消费水平的提升,消费者对水果的外观品质要求逐渐提高。目前,我国的水果分级主要依赖于人工与机械分级,存在分拣效率低、品质参差不齐的问题,很难满足消费者的需求。相对于传统的分级技术,机器视觉技术因具有高效、精确、非接触测量的优势,受到国内外学者的广泛关注。结合近年来国内外学者研究进展情况,对水果检测系统、图像去噪、图像分割技术以及特征提取做了详细介绍;对分类器识别算法进行了综合分析;阐述了基于机器视觉的水果分级研究现状及最新进展;并对未来可能发展趋势进行预测,为后续研究工作提供基础理论参考。

关键词: 水果, 机器视觉, 图像处理, 特征提取, 分级

Abstract:

With the development of economy and the improvement of consumption level, consumers have gradually increased the requirements for the appearance quality of fruits. At present, the classification of fruits in China mainly depends on manual and mechanical classification. There are problems of low sorting efficiency and uneven quality, which is difficult to meet consumer demand. Compared with traditional grading technology, machine vision technology has attracted widespread attention from scholars at home and abroad due to its advantages of high efficiency, accuracy, and non-contact measurement. Based on the research progress of scholars at home and abroad in recent years, this paper introduces fruit detection systems, image denoising, image segmentation techniques, and feature extraction in detail. It also comprehensively analyzes classifier recognition algorithms, and elaborates research on fruit classification based on machine vision current status and latest progress, forecasts possible future development trends, and provides basic theoretical references for subsequent research.

Key words: fruit, machine vision, image processing, feature extraction, classification