基于深度学习的种子质量检测综述

A review of seed quality detection based on deep learning

“种优则粮丰,粮安则民安”。我国是粮食大国,为保障粮食安全,需要紧抓种子质量,健康的种子和适宜的品种可以显著提高农作物的产量和质量。因此实现快速、无损地检测种子质量的技术对于粮食生产具有重要的意义。近年来,机器视觉作为一种无损检测方法在粮食种子质检领域得到了广泛应用。通过对粮食种子图像数据进行训练,最终得到能够准确、快速识别种子外观品质的模型。重点从数据集角度出发,总结了基于深度学习算法在小麦、稻米、玉米等主要粮食种子的外观品质检测方面的研究进展,并根据这些研究存在的不足对该领域未来的研究方向进行了展望。

Quality seeds lead to abundant grain production,which in turn ensures food security.As a major grain producer,China needs to prioritize seed quality to significantly improve crop yield and quality.Therefore,the development of fast and non-destructive techniques for seed quality assessment is of great significance in agricultural production.In recent years,machine vision has been widely used as a non-destructive method for grain seed quality inspection.By training models on grain seed image data,accurate and rapid identification of seed appearance quality can be achieved.This review focuses on the perspective of datasets and summarizes the research progress in the application of deep learning algorithms for appearance quality inspection of major grain seeds such as wheat,rice,and maize.The review also provides insights into future research directions based on the limitations identified in these studies.