Yang Haiying, Li Yanyu, Xin Liyong, Teng Shyh Wei, Pang Shaoning, Zhao Huiyi, Cao Yang, Zhou Xiaoguang
Department of Automation, Beijing University of Posts and Telecommunications, Beijing 100876, China.
Institute for Grain Storage & Logistics, Academy of National Food and Strategic Reserves Administration, Beijing 100037, China.
Foods. 2023 Oct 3;12(19):3653. doi: 10.3390/foods12193653.
Insect pests like and siblings are major threats to grain storage and processing, causing quality and quantity losses that endanger food security. These closely related species, having very similar morphological and biological characteristics, often exhibit variations in biology and pesticide resistance, complicating control efforts. Accurate pest species identification is essential for effective control, but workplace safety in the grain bin associated with grain deterioration, clumping, fumigator hazards, and air quality create challenges. Therefore, there is a pressing need for an online automated detection system. In this work, we enriched the stored-grain pest sibling image dataset, which includes 25,032 annotated samples of two species and five geographical strains from real warehouse and another 1774 from the lab. As previously demonstrated on the family, Convolutional Neural Networks demonstrate distinct advantages over other model architectures in detecting . Our CNN model, MCSNet+, integrates Soft-NMS for better recall in dense object detection, a Position-Sensitive Prediction Model to handle translation issues, and anchor parameter fine-tuning for improved matching and speed. This approach significantly enhances mean Average Precision (mAP) for and , reaching a minimum of 92.67 ± 1.74% and 94.27 ± 1.02%, respectively. Moreover, MCSNet+ exhibits significant improvements in prediction speed, advancing from 0.055 s/img to 0.133 s/img, and elevates the recognition rates of moving insect sibling species in real wheat storage and visible light, rising from 2.32% to 2.53%. The detection performance of the model on laboratory-captured images surpasses that of real storage facilities, with better results for compared to . Although inter-strain variances are less pronounced, the model achieves acceptable detection results across different geographical strains, with a minimum recognition rate of 82.64 ± 1.27%. In real-time monitoring videos of grain storage facilities with wheat backgrounds, the enhanced deep learning model based on Convolutional Neural Networks successfully detects and identifies closely related stored-grain pest images. This achievement provides a viable solution for establishing an online pest management system in real storage facilities.
像谷蠹及其近缘种这样的害虫是粮食储存和加工的主要威胁，会造成质量和数量损失，危及粮食安全。这些密切相关的物种具有非常相似的形态和生物学特征，但在生物学特性和抗药性方面常常表现出差异，这使得防治工作变得复杂。准确识别害虫种类对于有效防治至关重要，但粮堆中的工作场所安全存在粮食变质、结块、熏蒸剂危害和空气质量等问题，带来了挑战。因此，迫切需要一种在线自动检测系统。在这项工作中，我们扩充了储粮害虫近缘种图像数据集，该数据集包括来自真实仓库的两个物种和五个地理菌株的25032个带注释样本以及另外1774个实验室样本。如先前在象甲科上所证明的，卷积神经网络在检测谷蠹及其近缘种方面比其他模型架构具有明显优势。我们的卷积神经网络模型MCSNet+集成了Soft-NMS以在密集目标检测中获得更好的召回率，采用了位置敏感预测模型来处理平移问题，并对锚点参数进行微调以提高匹配度和速度。这种方法显著提高了谷蠹及其近缘种的平均精度均值（mAP），分别达到至少92.67±1.74%和94.27±1.02%。此外，MCSNet+在预测速度方面有显著提升，从0.055秒/图像提高到0.133秒/图像，并提高了在真实小麦储存环境和可见光下移动的谷蠹近缘种的识别率，从2.32%提高到2.53%。该模型在实验室采集图像上的检测性能优于真实储存设施，谷蠹的检测结果比米象更好。尽管不同地理菌株之间的差异不太明显，但该模型在不同谷蠹地理菌株上都取得了可接受的检测结果，最低识别率为82.64±1.27%。在以小麦为背景的粮食储存设施实时监测视频中，基于卷积神经网络的增强深度学习模型成功检测并识别了密切相关的储粮害虫图像。这一成果为在真实储存设施中建立在线害虫管理系统提供了可行的解决方案。
