No abstract

No abstract

Automatic machine vision-based defect detection has been successfully applied to many industrial visual inspection applications. However, automatic steel surface defect detection is still a challenging task due to diverse defect categories, low-contrast between defect and complex texture background. To address these challenges, a chained atrous spatial pyramid pooling network (CASPPNet) is proposed for steel surface defect detection. In CASPPNet, chained atrous spatial pyramid pooling is designed to enlarge receptive field and obtain enrich semantic information. An improved global attention feature fusion module is introduced to achieve feature interaction and salience. Moreover, residual boundary refinement block is introduced to get more complete defect boundary. Comparative experimental results verify that our method is superior to the state-of-the-art segmentation methods on public accessible SD-saliency-900 datasets and can meet the requirement of real-time online detection (the detection efficiency is at over 47 FPS on a single GPU).

Precise nitrogen (N) application ensures the best N status of potato plants to improve crop growth and food quality and to achieve the best N use efficiency. Four N fertilization levels (0, 2, 4 and 6 g N pot−1) were used to establish a critical N dilution curve (CNDC) of potato plants cultivated in substrates with a greenhouse environment. RGB images of potato plants were obtained, and a red–green fit index (RGFI) was calculated based on the linear relationship between R and G channels and the principle of the excess green index (EXG). The N in the substrate can meet the nutritional requirements of potato plants during the first 35 days after emergence. In order to solve the complex sampling problem of maintaining a sufficient N strip for aboveground dry biomass (DM) and crop nitrogen concentration, a reference curve method for detecting N status was proposed. RGFI and SPAD values from the economically optimum 4 g N pot−1 treatment were used to derive the reference curve. The RGFI and SPAD values from the 4 g N pot−1 treatment had high correlations and were fitted with a second-order polynomial function with an R2 value of 0.860 and an RMSE value of 2.10. The validation results show that the N concentration dilution curve constructed by RGFI and SPAD values can effectively distinguish N-limiting from non-N-limiting treatments, CNDCs constructed based on RGFI and SPAD values could be used as an effective N status monitoring tool for greenhouse potato production.

Due to complex environmental factors such as illumination, shading between leaves and fruits, shading between fruits, and so on, it is a challenging task to quickly identify red jujubes and count red jujubes in orchards. A counting method of red jujube based on improved YOLOv5s was proposed, which realized the fast and accurate detection of red jujubes and reduced the model scale and estimation error. ShuffleNet V2 was used as the backbone of the model to improve model detection ability and light the weight. In addition, the Stem, a novel data loading module, was proposed to prevent the loss of information due to the change in feature map size. PANet was replaced by BiFPN to enhance the model feature fusion capability and improve the model accuracy. Finally, the improved YOLOv5s detection model was used to count red jujubes. The experimental results showed that the overall performance of the improved model was better than that of YOLOv5s. Compared with the YOLOv5s, the improved model was 6.25% and 8.33% of the original network in terms of the number of model parameters and model size, and the Precision, Recall, F1-score, AP, and Fps were improved by 4.3%, 2.0%, 3.1%, 0.6%, and 3.6%, respectively. In addition, RMSE and MAPE decreased by 20.87% and 5.18%, respectively. Therefore, the improved model has advantages in memory occupation and recognition accuracy, and the method provides a basis for the estimation of red jujube yield by vision.