基于机器视觉的猴头菇品质快速无损检测与分级

传统的猴头菇品质检测与分级主要依靠人工分拣来完成，其主观性强、精度相差大、效率低，浪费了大量人力物力资源。鉴于此，为了实现猴头菇的快速无损等级评估，该研究引入机器视觉技术，提出了一种猴头菇品质的快速无损检测与分级方法，设计一套基于机器视觉的猴头菇品质快速无损检测与智能分级设备，并通过图像处理和软件设计开发一套猴头菇智能快速无损检测分级系统。通过加色法混色模型（RGB）对猴头菇的颜色特征的快速检测与等级的判定；采用图像阈值分割和Canny边缘检测，实现猴头菇完整度的判定；使用最小外接圆法对猴头菇的大小进行实时计算，完成猴头菇直径大小的判别；基于Microsoft Visual Studio 2017平台开发一套猴头菇品质快速无损检测可视化平台。试验证明，基于机器视觉的猴头菇品质快速无损检测与分级系统检测准确率达到97.07%，速度达到人工的5倍多。验证了系统的可靠性和可行性，为食品工业的智能化生产和加工提供了技术支撑，推动了机器视觉技术在食品行业的应用。

Rapid Non-destructive Testing and Grading of Hericium erinaceus Based on Machine Vision

The traditional quality inspection and classification of the edible mushroom Hericium erinaceus mainly depend on manual sorting, a process that is highly subjective and inefficient, resulting in uneven accuracy and significant waste of human and material resources. In order to realize the rapid non-destructive grade evaluation of H. erinaceus, we incorporated machine vision technology (image processing and software design) into the sorting and grading process. The color characteristics and grade of H. erinaceus were quickly detected by applying the additive color mixing model (RGB). Image threshold segmentation and Canny edge detection were used to determine the integrity of the material, and the minimum circumscribed circle method was used to calculate the size of the sample. A visual platform for rapid non-destructive testing of Hericium erinaceus quality was developed based on Microsoft Visual Studio 2017 platform. The results of these test confirmed the accuracy (97.07%) of the rapid non-destructive testing and grading system of Hericium erinaceus quality based on machine vision, and the process speed was more than five times that of the usual manual process. The reliability and feasibility of the system is verified, which should lead to further development of machine vision technology in the food processing industries.