The accurate acquisition of safflower filament information is the prerequisite for robotic picking operations. To detect safflower filaments accurately in different illumination, branch and leaf occlusion, and weather conditions, an improved Faster R-CNN model for filaments was proposed. Due to the characteristics of safflower filaments being dense and small in the safflower images, the model selected ResNeSt-101 with residual network structure as the backbone feature extraction network to enhance the expressive power of extracted features. Then, using Region of Interest (ROI) Align improved ROI Pooling to reduce the feature errors caused by double quantization. In addition, employing the partitioning around medoids (PAM) clustering was chosen to optimize the scale and number of initial anchors of the network to improve the detection accuracy of small-sized safflower filaments. The test results showed that the mean Average Precision (mAP) of the improved Faster R-CNN reached 91.49%. Comparing with Faster R-CNN, YOLOv3, YOLOv4, YOLOv5, and YOLOv6, the improved Faster R-CNN increased the mAP by 9.52%, 2.49%, 5.95%, 3.56%, and 1.47%, respectively. The mAP of safflower filaments detection was higher than 91% on a sunny, cloudy, and overcast day, in sunlight, backlight, branch and leaf occlusion, and dense occlusion. The improved Faster R-CNN can accurately realize the detection of safflower filaments in natural environments. It can provide technical support for the recognition of small-sized crops.

Discrete element modeling and parameter calibration of safflower biomechanical properties

Development of a rotary push-cut-type negative-pressure-airflow end-effector for harvesting safflower filaments

To improve the stubble avoidance rate of no-till wheat seeders under the environment of corn stubble, the side-shift stubble avoidance system was developed in the present study based on deviation-perception fusion technology. Firstly, the amount of implement deviation compensation was accurately obtained by the information sensing system. Then, a fuzzy PID algorithm was proposed to optimize the performance of the control system. Finally, the simulation model of the control system was developed by Matlab/Simulink. The simulation results showed that the regulation time required for the system response curve to reach the steady state was 1.0 s, and the maximum overshoot was 8.2%. To verify the performance of the side-shift stubble avoidance system, an orthogonal test was conducted. The optimal combination of parameters was obtained, including the operation speed of 1.3 m/s, the hydraulic cylinder speed of 25 mm/s, and the straw mulch amount of 1.2 kg/m². The field verification test was performed under the optimal parameters, showing the stubble avoidance rate was 90.6%, and the mean value of the sowing variation coefficient was 27.12%, which fully satisfied the requirements of stubble avoidance and sowing operation.

Clamping root-cutting end-effector for harvesting fresh safflower