A field-based apple detection and grading device was developed and used to detect and grade apples in the field using a deep learning framework. Four features were selected for apple grading, namely, size, color, shape, and surface defects, and detection algorithms were designed to discriminate between the four features using machine vision and other methods. Then, the four apple features were fused, and a support vector machine (SVM) was used for infield apple grading into three grades: first-grade fruit, second-grade fruit, and other-grade fruit. The results showed that for a single index, the accuracy of detecting the apple size, the fruit shape, the color, and the surface defects, were 99.04%, 97.71%, 98%, and 95.85%. The grading accuracies for the first-grade fruit, second-grade fruit, other-grade fruit, and the average grading accuracy based on multiple features were 94.55%, 95.71%, 100%, and 95.49%, respectively. The field experiment showed that the average grading accuracy was 94.12% when the feeding interval of the apples was less than 1.5 s and the walking speed did not exceed 0.5 m/s, meeting the accuracy requirements of field-based apple grading.

The current mechanical harvesting methods of Lycium barbarum L. are labor intensive and cause too much damage, but vibrating and comb-brushing harvesting can increase the efficiency while minimizing the damage. However, optimizing the main factors and their parameter values of vibrating and comb-brushing harvesting is challenging. To achieve the high-efficiency and low-damage harvesting of L. barbarum, firstly, the mechanical models of the materials used in the experiments were established based on the physical tests. Then, the vibrating and comb-brushing harvesting simulations were conducted based on FEM to acquire the ranges of the parameter values. The effects of the rotating speed, material, and amplitude on the harvesting rate of ripe fruit and harvesting rate of unripe fruit, as well as the damage rate of ripe fruit were determined based on RSM. Finally, the optimized parameters were obtained and verified using the field experiments. The field experiments showed that the harvesting rate of ripe fruit was 85.8%, the harvesting rate of unripe fruit was 10.5%, and the damage rate of ripe fruit was 9.7%. The findings provided the optimal parameter values, which were a design basis for the vibrating and comb-brushing harvesters of L. barbarum.

In a previously published study, to optimize the vibrating and comb-brushing harvesting, the main factors and their parameter values were obtained based on the FEM and RSM. However, the study was based on the extensive cultivation mode which need to be improve. To realize the mechanization of the harvesting of Lycium barbarum L., as well as to face the standardized hedge cultivation mode, a vibrating and comb-brushing harvester machine was designed, which was primarily composed of an execution system, a motion system, and a control system. The mathematical model between the harvest effect index and the operation parameters was established based on response surface methodology (RSM). The effects of various parameters on the harvest index were analyzed, and the best parameter combination was determined: a vibration frequency of 38.73 Hz, a brush speed of 14.21 mm/s, and an insertion depth of 26.07 mm. The field experiment showed that the harvesting rate of ripe fruit was 83.65%, the harvesting rate of unripe fruit was 7.22%, the damage rate of the ripe fruit was 11.49%, and the comprehensive picking index was 87.85. The findings provided a reference for the development of L. barbarum harvesting mechanization in a standardized hedge cultivation mode.