Crop lodging is an important cause of direct economic losses and secondary disease transmission in agricultural production. Most existing methods for segmenting wheat lodging areas use a large-volume network, which poses great difficulties for annotation and crop monitoring in real time. Therefore, an ultra-lightweight model, Lodging-U2NetP (L-U2NetP), based on a novel annotation strategy which crops the images before annotating them (Crop-annotation), was proposed and applied to RGB images of wheat captured with an unmanned aerial vehicle (UAV) at a height of 30 m during the maturity stage. In the L-U2NetP, the Dual Cross-Attention (DCA) module was firstly introduced into each small U-structure effectively to address semantic gaps. Then, Crisscross Attention (CCA) was used to replace several bulky modules for a stronger feature extraction ability. Finally, the model was compared with several classic networks. The results showed that the L-U2NetP yielded an accuracy, F1 score, and IoU (Intersection over Union) for segmenting of 95.45%, 93.11%, 89.15% and 89.72%, 79.95%, 70.24% on the simple and difficult sub-sets of the dataset (CA set) obtained using the Crop-annotation strategy, respectively. Additionally, the L-U2NetP also demonstrated strong robustness in the real-time detection simulations and the dataset (AC set) obtained using the mainstream annotation strategy, which annotates images before cropping (Annotation-crop). The results indicated that L-U2NetP could effectively extract wheat lodging and the Crop-annotation strategy provided a reliable performance which is comparable with that of the mainstream one.

Fusarium head blight (FHB) is a major threat to global wheat production. Recent reviews of wheat FHB focused on pathology or comprehensive prevention and lacked a summary of advanced detection techniques. Unlike traditional detection and management methods, wheat FHB detection based on various imaging technologies has the obvious advantages of a high degree of automation and efficiency. With the rapid development of computer vision and deep learning technology, the number of related research has grown explosively in recent years. This review begins with an overview of wheat FHB epidemic mechanisms and changes in the characteristics of infected wheat. On this basis, the imaging scales are divided into microscopic, medium, submacroscopic, and macroscopic scales. Then, we outline the recent relevant articles, algorithms, and methodologies about wheat FHB from disease detection to qualitative analysis and summarize the potential difficulties in the practicalization of the corresponding technology. This paper could provide researchers with more targeted technical support and breakthrough directions. Additionally, this paper provides an overview of the ideal application mode of the FHB detection technologies based on multi-scale imaging and then examines the development trend of the all-scale detection system, which paved the way for the fusion of non-destructive detection technologies of wheat FHB based on multi-scale imaging.

Traditional detection methods of crop information are often destructive and low efficiency. In this study, a new evaluation method was developed based on photosynthesis and chlorophyll fluorescence. Via analysis of the changes under NaCl stress during the seedling stage of two varieties, the salt resistance mechanism of soybeans was explored and a non-destructive stress-recognition method was developed. In this experiment, two soybean varieties were treated with one of four levels of NaCl stress: CK (0 mmol/L), LS (50 mmol/L), MS (100 mmol/L), and HS (150 mmol/L), for 15 d. The normal functions of the photosynthetic system of soybeans were enhanced under LS NaCl stress, but were inhibited under HS NaCl stress. Biomass, net leaf photosynthetic rate (Pn), stomatal conductance (gs), intercellular carbon dioxide concentration (Ci), transpiration rate (Tr), chlorophyll fluorescence parameters Y(II) and PSII decreased. However, in contrast to the findings of other studies on the influence of severe drought stress on soybean for long periods in which non-photochemical quenching coefficient (qN) decreased, this parameter increased under salt stress in soybean. The results demonstrate that the method developed is a promising tool for rapid and non-destructive detection of soybean photosynthetic responses under salt stress in the field. Keywords: biomass, chlorophyll fluorescence parameters, NaCl stress, photosynthetic parameters, soybean DOI: 10.25165/j.ijabe.20211403.5941 Citation: Luo B, Wang C, Wang X D, Zhang H, Zhou Y N, Wang W S, et al. Changes in photosynthesis and chlorophyll fluorescence in two soybean (Glycine max) varieties under NaCl stress. Int J Agric & Biol Eng, 2021; 14(3): 76–82.

Rapid and non-destructive monitoring of Vitamin C (VC) content in lettuce across different growth stages is essential for intelligent precision harvesting and post-harvest quality management. This study aimed to develop a highly accurate and robust model for predicting VC content in lettuce using near-infrared (NIR) spectroscopy. Spectral data in the 1350-1800 nm range were collected from 326 lettuce samples at the critical stages-191 from the seedling stage and 135 from the maturity stage, along with their VC values determined by chemical methods. Stratified random sampling selected total 97 samples from seedling and maturity stages as an independent external validation set for generalization assessment. Normalization correction (NC) and direct orthogonal signal correction (DOSC) were applied for preprocessing to remove background and noise interference. Three effective wavelengths (EWs) algorithms and one spectral feature extraction algorithm were employed for model optimization comparison, with the deep belief network (DBN) method yielding the best results.Five regression models were constructed for comparison, the improved whale optimization algorithm-optimized least squares support vector machine (IWOA-LSSVM) model exhibited superior predictive performance. The results show that the optimal model yielded overall Rp2 of 0.9252, RMSE of 0.3712 and MAE of 0.2990 on the external set, while for seedling and maturity stages the metrics were 0.8800, 0.4608, and 0.4027 and 0.9500, 0.3341, and 0.2436, respectively. These results demonstrated that the VC content detection model of lettuce, founded on DBN-IWOA-LSSVM, provided a modelling foundation for internal quality detection of lettuce and other vegetables.