It is a challenging problem to detect the apple in natural environment using traditional object recognition algorithms due to occlusion, fluctuating illumination and complex backgrounds. Deep learning methods for object detection make impressive progress, which can automatically extract the number, pixel position, size and other features of apples from the images. In this paper, four deep learning recognition models, Faster RCNN based on AlexNet, Faster RCNN based on ResNet101, YOLOv3 based on DarkNet53 and improved YOLOv3 were employed to carry out recognition experiments on red and green apple under three illumination and two image sharpness conditions, with the transfer learning to accelerate the training process. The results showed that improved YOLOv3 model had the best recognition effect among the four detection models. F1 value of red apple recognition was 95.0%, 94.6% and 94.1% for normal, insufficient and excessive illumination, respectively, and F1 value of green apple recognition was 94.9%, 94.0% and 91.1%. There were F1 value of 92.8% and 92.1% for red and green apple recognition in blurred images, respectively. Moreover, improved YOLOv3 algorithm still had the better performance for occlusion, spot, overlap and incomplete apples, with a recognition recall rate higher than 88.5%. It can be concluded that improved YOLOv3 algorithm can provide a more efficient way for apple detection in natural environment.

To address the lightweight and real-time issues of coal sorting detection, an intelligent detection method for coal and gangue, Our-v8, was proposed based on improved YOLOv8. Images of coal and gangue with different densities under two diverse lighting environments were collected. Then the Laplacian image enhancement algorithm was proposed to improve the training data quality, sharpening contours and boosting feature extraction; the CBAM attention mechanism was introduced to prioritize crucial features, enhancing more accurate feature extraction ability; and the EIOU loss function was added to refine box regression, further improving detection accuracy. The experimental results showed that Our-v8 for detecting coal and gangue in a halogen lamp lighting environment achieved excellent performance with a mean average precision (mAP) of 99.5%, was lightweight with FLOPs of 29.7, Param of 12.8, and a size of only 22.1 MB. Additionally, Our-v8 can provide accurate location information for coal and gangue, making it ideal for real-time coal sorting applications.

The objective is to develop a portable device capable of promptly identifying root rot in the field. This study employs hyperspectral imaging technology to detect root rot by analyzing spectral variations in chili pepper leaves during times of health, incubation, and disease under the stress of root rot. Two types of chili pepper seeds (Manshanhong and Shanjiao No. 4) were cultured until they had grown two to three pairs of true leaves. Subsequently, robust young plants were infected with Fusarium root rot fungi by the root-irrigation technique. The effective wavelength for discriminating between distinct stages was determined using the successive projections algorithm (SPA) after capturing hyperspectral images. The optimal index related to root rot between each normalized difference spectral index (NDSI) was obtained using the Pearson correlation coefficient. The early detection of root rot illness can be modeled using spectral information at effective wavelengths and in NDSI, together with the application of partial least squares discriminant analysis (PLS-DA), least squares support vector machine (LSSVM), and back-propagation (BP) neural network technology. The SPA-BP model demonstrates outstanding predictive capabilities compared with other models, with a classification accuracy of 92.3% for the prediction set. However, employing SPA to acquire an excessive number of efficient wave-lengths is not advantageous for immediate detection in practical field scenarios. In contrast, the NDSI (R₄₄₅, R₄₃₃)-BP model uses only two wavelengths of spectral information, but the prediction accuracy can reach 89.7%, which is more suitable for rapid detection of root rot. This thesis can provide theoretical support for the early detection of chili root rot and technical support for the design of a portable root rot detector.