Rice seedling segmentation is a fundamental process of extracting the guidance line for automated rice transplanters with a visual navigation system, which can provide crop row information to ensure the transplanter plants seedlings along the crop row without damaging seedlings. However, obtaining accurate rice seedling segmentation in paddy fields is still a challenging task. In this paper, a rice seedling segmentation method in paddy fields is proposed. The method mainly consists of two steps: image graying and threshold segmentation. In the procedure of image graying, the RGB (Red Green Blue) seedling image is first converted into the YCrCb color space and a Cg component is constructed. A color-index 2Cg-Cb-Cr is then constructed for image graying based on the excess green index (2G-R-B), which can reduce the influence of illumination variation on the equality of image graying. For the second step, an improved Otsu method is proposed to segment rice seedlings. With respect to the improved Otsu method in this research, the background variance of within class variance is weighted by a probability parameter to ensure that the method works well for both bimodal and near-unimodal histogram images, and the search range of gray levels is constrained to reduce the time to search the segmentation threshold. Experimental results indicate that the proposed method achieves better segmentation results and reduces the computational cost compared with the traditional Otsu method and other improved Otsu methods.

Crop segmentation is a fundamental step of extracting the guidance line for an automated agricultural machine with a visual navigation system. However, the segmentation quality of green crop is subject to the influence of outdoor illumination. To improve the visual effect of crop images and accuracy of crop segmentation under complex illumination environments, an index-based crop segmentation method with luminance partition correction and adaptive thresholding is proposed in this paper. The method receives the luminance component from the given RGB image and employs two adaptive thresholds to segment the luminance image into the dark, normal and bright areas. Then, a partition Gamma functions are adaptively selected to improve the brightness levels of the dark and bright regions in which the Gamma correction parameter is adaptively determined by the local distribution characteristics of illumination, and the corrected image is locally fused and transformed to the RGB space with color saturation restoration. Finally, the method performs pre-segmentation using the ExG (excess green index) color index and an Otsu threshold and uses this pre-segmentation result to calculate the segmentation threshold for the final segmentation. Experimental results show that the proposed luminance partition correction approach can effectively increase the brightness level of the dark region, and decrease the brightness levels in the bright region. In addition, the proposed method can achieve better segmentation results compared with the traditional Otsu method employed in before and after luminance correction.

In order to realize automatic steering controls of rice transplanters in paddy fields, an automatic steering control algorithm is essential. In this study, combining the fuzzy control with the proportional-integral-derivative (PID) control and the kinematics model, a compound fuzzy PID controller was proposed to adjust the real time data of the PID parameters for the automatic steering control. The Kubota SPU-68C rice transplanter was then modified with the new controller. Next, an automatic steering control experimental with the modified transplanter was carried out under two conditions of linear tracking and headland turning in verifying the automatic steering effect of the transplanter in different steering angle situations. The results showed that the deviation with the new controller and the modified transplanter was acceptable, with maximum deviation in linear tracking of 7.5 cm, the maximum headland turning a deviation of 11.5 cm, and the average a deviation of less than 5 cm. In conclusion, within the allowable deviation range of the field operation of the rice transplanter, the proposed algorithm successfully realized automatic steering controls of the transplanter under different steering angles.

The identification of the growth and development period of rice is of great significance to achieve high-yield and high-quality rice. However, the acquisition of rice growth period information mainly relies on manual observation, which has problems such as low efficiency and strong subjectivity. In order to solve these problems, a lightweight recognition method is proposed to automatically identify the growth period of rice: Small-YOLOv5, which is based on improved YOLOv5s. Firstly, the new backbone feature extraction network MobileNetV3 was used to replace the YOLOv5s backbone network to reduce the model size and the number of model parameters, thus improving the detection speed of the model. Secondly, in the feature fusion stage of YOLOv5s, we introduced a more lightweight convolution method, GsConv, to replace the standard convolution. The computational cost of GsConv is about 60–70% of the standard convolution, but its contribution to the model learning ability is no less than that of the standard convolution. Based on GsConv, we built a lightweight neck network to reduce the complexity of the network model while maintaining accuracy. To verify the performance of Small-YOLOv5s, we tested it on a self-built dataset of rice growth period. The results show that compared with YOLOv5s (5.0) on the self-built dataset, the number of the model parameter was reduced by 82.4%, GFLOPS decreased by 85.9%, and the volume reduced by 86.0%. The mAP (0.5) value of the improved model was 98.7%, only 0.8% lower than that of the original YOLOv5s model. Compared with the mainstream lightweight model YOLOV5s- MobileNetV3-Small, the number of the model parameter was decreased by 10.0%, the volume reduced by 9.6%, and the mAP (0.5:0.95) improved by 5.0%—reaching 94.7%—and the recall rate improved by 1.5%—reaching 98.9%. Based on experimental comparisons, the effectiveness and superiority of the model have been verified.