The accurate identification of citrus fruits is important for fruit yield estimation in complex citrus orchards. In this study, the YOLOv7-tiny-BVP network is constructed based on the YOLOv7-tiny network, with citrus fruits as the research object. This network introduces a BiFormer bilevel routing attention mechanism, which replaces regular convolution with GSConv, adds the VoVGSCSP module to the neck network, and replaces the simplified efficient layer aggregation network (ELAN) with partial convolution (PConv) in the backbone network. The improved model significantly reduces the number of model parameters and the model inference time, while maintaining the network’s high recognition rate for citrus fruits. The results showed that the fruit recognition accuracy of the modified model was 97.9% on the test dataset. Compared with the YOLOv7-tiny, the number of parameters and the size of the improved network were reduced by 38.47% and 4.6 MB, respectively. Moreover, the recognition accuracy, frames per second (FPS), and F1 score improved by 0.9, 2.02, and 1%, respectively. The network model proposed in this paper has an accuracy of 97.9% even after the parameters are reduced by 38.47%, and the model size is only 7.7 MB, which provides a new idea for the development of a lightweight target detection model.

Rapid and nondestructive prediction of chlorophyll content and response to the growth of various crops using remote sensing technology is a prominent topic in agricultural remote sensing research. Bordeaux mixture has been extensively employed for managing citrus diseases, such as black star and ulcer disease. However, the presence of pesticide residues in Bordeaux mixture can significantly modify the spectral response of the citrus canopy, thereby exerting a substantial influence on the accurate prediction of agronomic indices in fruit trees. In this study, we used unmanned aerial vehicle (UAV) multispectral imaging technology to obtain remote sensing imagery of Bordeaux-covered citrus canopies during the months of July, September, and November. We integrated spectral and texture information to construct a high-dimensional feature dataset and performed data downscaling and feature optimization. Furthermore, we established four machine learning models, namely, partial least squares regression (PLS), ridge regression (RR), ridge, random forest (RF), and support vector regression (SVR). Our objectives were to identify the most effective prediction model for estimating the SPAD (soil plant analysis development) value of Bordeaux-covered citrus canopies, assess the variation in prediction accuracy between fused features and individual features, and investigate the impact of Bordeaux solution on the spectral reflectance of the citrus canopy. The results showed that (1) the impact of Bordeaux mixture on citrus canopy reflectance bands ranked from the highest to the lowest as follows: near-infrared band at 840 nm, red-edge band at 730 nm, blue band at 450 nm, green band at 560 nm, and red band at 650 nm. (2) Fused feature models had better prediction ability than single-feature modeling, with an average R² value of 0.641 for the four model test sets, improving by 0.117 and 0.039, respectively, compared with single-TF (texture feature) and -VI (vegetation index) modeling, and the test-set root-mean-square error (RMSE) was 2.594 on average, which was 0.533 and 0.264 lower than single-TF and -VI modeling, respectively. (3) Multiperiod data fusion effectively enhanced the correlation between features and SPAD values and consequently improved model prediction accuracy. Compared with accuracy based on individual months, R improved by 0.013 and 0.011, while RMSE decreased by 0.112 and 0.305. (4) The SVR model demonstrated the best performance in predicting citrus canopy SPAD under Bordeaux solution coverage, with R² values of 0.629 and 0.658, and RMSE values of 2.722 and 2.752 for the training and test sets, respectively.

Early crop yield prediction is a major challenge in precision agriculture, and efficient and rapid yield prediction is highly important for sustainable fruit production. The accurate detection of major fruit characteristics, including flowering, green fruiting, and ripening stages, is crucial for early yield estimation. Currently, most crop yield estimation studies based on the YOLO model are only conducted during a single stage of maturity. Combining multi-growth period data for crop analysis is of great significance for crop growth detection and early yield estimation. In this study, a new network model, YOLOv8-RL, was proposed using citrus multigrowth period characteristics as a data source. A citrus yield estimation model was constructed and validated by combining network identification counts with manual field counts. Compared with YOLOv8, the number of parameters of the improved network is reduced by 50.7%, the number of floating-point operations is decreased by 49.4%, and the size of the model is only 3.2 MB. In the test set, the average recognition rate of citrus flowers, green fruits, and orange fruits was 95.6%, the mAP@.5 was 94.6%, the FPS value was 123.1, and the inference time was only 2.3 milliseconds. This provides a reference for the design of lightweight networks and offers the possibility of deployment on embedded devices with limited computational resources. The two estimation models constructed on the basis of the new network had coefficients of determination R² values of 0.91992 and 0.95639, respectively, with a prediction error rate of 6.96% for citrus green fruits and an average error rate of 3.71% for orange fruits. Compared with network counting, the yield estimation model had a low error rate and high accuracy, which provided a theoretical basis and technical support for the early prediction of fruit yield in complex environments.