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.

To accurately and efficiently remove unripe fruit, flowers, leaves, and other impurities in machine-harvested Lycium barbarum L., winnowing equipment for machine-harvested L. barbarum based on the principle that different materials have different flight coefficients was designed. To optimize the structure and working parameters of winnowing equipment, this study adopted the free flow resistance model to establish a horizontal airflow model based on C++ in Microsoft Visual Studio. A discrete element method (DEM) simulation of ripe fruit in the horizontal airflow was performed using EDEM software. Results showed that the optimal parameters included an airflow speed of 5-6 m/s, input conveyor speed of 0.4-0.6 m/s, and input-output conveyor distance of 260-270 mm. We used three factors and three levels in a quadratic orthogonal rotation design to establish mathematical models regarding the rate of impurity change and the clearance rate of ripe fruit based on the airflow speed, input conveyor speed, and input-output conveyor distance. We also analyzed the effects of all factors on the rate of impurity change and the clearance rate of ripe fruit. The optimal parameter combination was an airflow speed of 5.52 m/s, input conveyor speed of 0.5 m/s, and input-output conveyor distance of 265.04 mm. The field experiment showed that the rate of impurity change and the clearance rate of ripe fruit were 89.74% and 8.71%, respectively. Findings provide a design basis for future research on winnowing equipment for machine-harvested L. barbarum.