Weeding is a key link in agricultural production. Intelligent mechanical weeding is recognized as environmentally friendly, and it profoundly alleviates labor intensity compared with manual hand weeding. While intelligent mechanical weeding can be implemented only when a large number of disciplines are intersected and integrated. This article reviewed two important aspects of intelligent mechanical weeding. The first one was detection technology for crops and weeds. The contact sensors, non-contact sensors and machine vision play pivotal roles in supporting crop detection, which are used for guiding the movements of mechanical weeding executive parts. The second one was mechanical weeding executive part, which include hoes, spring teeth, fingers, brushes, swing and rotational executive parts, these parts were created to adapt to different soil conditions and crop agronomy. It is a fact that intelligent mechanical weeding is not widely applied yet, this review also analyzed the related reasons. We found that compared with the biochemical sprayer, intelligent mechanical weeding has two inevitable limitations: The higher technology cost and lower working efficiency. And some conclusions were commented objectively in the end.

In Northeast China, the maize (Zea mays L.) harvester header height is still manually controlled, and the control precision is poor, which limits the quality of maize harvesting and is unfavorable to the long-term development of agricultural harvesting. This work created an adaptive adjustment method for header height to address this issue. A maize harvester header, an STM32 control unit and key module, a hydraulic adjustment mechanism, and a pressure-wheel profiling device comprise the majority of the system. In this scenario, the proposed pressure-wheel profiling device is mounted to the ridge’s surface and walks along it, delivering real-time data on terrain changes. The terrain change signals are received and processed in real time by the STM32 control unit, which subsequently operates the hydraulic adjustment mechanism to alter the header height. The structural dimensions and operational parameters of the essential components for pressure-wheel profiling device were determined using force analysis. A kinematic analysis of the hydraulic adjusting mechanism was performed, and the expression of the relationship between the displacement of hydraulic cylinder and the header height was obtained. The pressure-wheel profiling device was calibrated, the adaptive adjustment parameter model was constructed, and the PID control technique was employed to achieve automatic header height adjustment. This study analyzed the effects of harvester different forward speeds (1.25, 1.45, 1.65, 1.85, and 2.05 m/s) and different operating modes (open and unopened system) on the test indexes (Errors in the automatic adjustment of header height, Cob loss rate, Coefficient of variation in stubble height). The results revealed that, with the adaptive adjustment system mode on, the average difference between the measured actual header height and the predetermined height was 9.96 mm, the average coefficient of variation in stubble height was lowered by 34.44%, and the average cob loss rate was decreased by 7.98%, both of which may accommodate the needs of maize harvester header height adjustment. This study serves as a reference for the automated design of a maize harvester for monopoly crops.

The accurate spraying of herbicides and intelligent mechanical weeding operations are the main ways to reduce the use of chemical pesticides in fields and achieve sustainable agricultural development, and an important prerequisite for achieving these is to identify field crops and weeds accurately and quickly. To this end, a semantic segmentation model based on an improved U-Net is proposed in this paper to address the issue of efficient and accurate identification of vegetable crops and weeds. First, the simplified visual group geometry 16 (VGG16) network is used as the coding network of the improved model, and then, the input images are continuously and naturally down-sampled using the average pooling layer to create feature maps of various sizes, and these feature maps are laterally integrated from the network into the coding network of the improved model. Then, the number of convolutional layers of the decoding network of the model is cut and the efficient channel attention (ECA) is introduced before the feature fusion of the decoding network, so that the feature maps from the jump connection in the encoding network and the up-sampled feature maps in the decoding network pass through the ECA module together before feature fusion. Finally, the study uses the obtained Chinese cabbage and weed images as a dataset to compare the improved model with the original U-Net model and the current commonly used semantic segmentation models PSPNet and DeepLab V3+. The results show that the mean intersection over union and mean pixel accuracy of the improved model increased in comparison to the original U-Net model by 1.41 and 0.72 percentage points, respectively, to 88.96% and 93.05%, and the processing time of a single image increased by 9.36 percentage points to 64.85 ms. In addition, the improved model in this paper has a more accurate segmentation effect on weeds that are close to and overlap with crops compared to the other three comparison models, which is a necessary condition for accurate spraying and accurate weeding. As a result, the improved model in this paper can offer strong technical support for the development of intelligent spraying robots and intelligent weeding robots.