The present study aimed to investigate the effect of rotary microwave vacuum drying (RMVD), radio frequency vacuum drying (RFVD), vacuum far infrared drying (VFID), vacuum drying (VD), hot air drying (HD) and natural drying (ND) on the drying characteristics, active ingredients and microstructure of Codonopsis pilosulae slices. Compared with the fitting results of the four models, the Weibull model is the most suitable drying model for Codonopsis. The RFVD and HD color difference values were smaller compared to ND. The effective moisture diffusivity (Deff) under different drying methods was between 0.06 × 10⁻⁸ m²/s and 3.95 × 10⁻⁸ m²/s. RMVD-dried products had the shortest drying time and retained more active ingredients. The microstructure analysis revealed that the porous structure of RMVD is more favorable for water migration. RMVD is a promising dehydration method for obtaining high-value-added dried Codonopsis products.

To overcome the problems of a long conventional drying time, low energy efficiency, and poor product quality, a segmented drying approach was developed for fresh wolfberry (Lycium barbarum) using a radio frequency (RF)-hot air drying process, which was investigated under different parameters of plate spacing (80, 90, 100 mm), vacuum degree (0.015, 0.025, 0.035 Mpa), and hot air temperature (50, 55, 60 °C). Analysis of the wolfberry’s drying characteristics, comprehensive quality, and microstructure indicated that: combined drying was faster and less time-consuming than natural drying or hot air drying, and components such as polysaccharides, ascorbic acid, and betaine in wolfberries were effectively retained. Based on the acceptable drying rate, stable temperature application, and avoidance of arcing effects, the optimal combined segmented drying parameters were determined to be as follows: a plate spacing of 90 mm, vacuum degree of 0.025 MPa, and air temperature of 55 °C. For the dried wolfberries under these conditions, the total drying time was 17 h and the berries had an improved comprehensive quality, the content of total soluble sugars was 0.62 g/g, total phenol was 10.01 mg/g, total flavonoids was 2.60 mg/g, V_C was 3.18 mg/100 g, betaine was 3.48%, oxidation resistance represented by an inhibition rate was 66.14%, color was better, and rehydration rate was 48.56%. The microstructure was more regular because of the special dielectric heating characteristics of RF vacuuming. Despite the differing drying characteristics of individual materials, the overall RF-hot air combined drying process was found to achieve high-quality dehydration of wolfberries.

In response to the issues of missed detection, false positives, and low recognition rates for specific weed species during weed detection, a YOLOv8-based improved weed detection algorithm named EDS-YOLOv8 is proposed. Improvements were made in three main aspects. First, the YOLOv8 backbone network was enhanced with EfficientViT and RepViT architectures to improve the detection capability of dense-type weeds. Second, different attention mechanisms were added, such as SimAM and EMA, to learn 3D weights and achieve full fusion of features. BiFormer was introduced for dynamic sparse attention and resource allocation. Third, significant module improvement involved introducing dynamic snake convolution into the C2f module to further enhance detection capabilities for deformable objects, especially needle-shaped weeds. The improved model is validated on the established weed dataset. The results show that combining the original backbone network with dynamic snake convolutions yields the highest performance improvement. Precision, recall, mAP (0.5), and mAP (0.5:0.95) are improved by 5.6%, 5.8%, 6.4%, and 1%, respectively, and ablation experiments on the effects of the three improvement methods on model performance show that using EfficientViT as the backbone network while simultaneously improving the crucial module and adding the SimAM attention mechanism effectively enhances the model’s performance. Precision, recall, mAP (0.5), and mAP (0.5:0.95) are improved by 6%, 5.9%, 6.4%, and 0.7%, respectively.

The target recognition algorithm is one of the core technologies of Zanthoxylum pepper-picking robots. However, most existing detection algorithms cannot effectively detect Zanthoxylum fruit covered by branches, leaves and other fruits in natural scenes. To improve the work efficiency and adaptability of the Zanthoxylum-picking robot in natural environments, and to recognize and detect fruits in complex environments under different lighting conditions, this paper presents a Zanthoxylum-picking-robot target detection method based on improved YOLOv5s. Firstly, an improved CBF module based on the CBH module in the backbone is raised to improve the detection accuracy. Secondly, the Specter module based on CBF is presented to replace the bottleneck CSP module, which improves the speed of detection with a lightweight structure. Finally, the Zanthoxylum fruit algorithm is checked by the improved YOLOv5 framework, and the differences in detection between YOLOv3, YOLOv4 and YOLOv5 are analyzed and evaluated. Through these improvements, the recall rate, recognition accuracy and mAP of the YOLOv5s are 4.19%, 28.7% and 14.8% higher than those of the original YOLOv5s, YOLOv3 and YOLOv4 models, respectively. Furthermore, the model is transferred to the computing platform of the robot with the cutting-edge NVIDIA Jetson TX2 device. Several experiments are implemented on the TX2, yielding an average time of inference of 0.072, with an average GPU load in 30 s of 20.11%. This method can provide technical support for pepper-picking robots to detect multiple pepper fruits in real time.

In order to improve the shortcomings of uneven heating of traditional microwave drying and to maximally maintain food quality after harvest, a rotary microwave vacuum drying equipment was fabricated and used for drying experiments on Angelica sinensis to explore the effects of drying temperature, slice thickness, and vacuum degree on drying characteristics, physicochemical quality, and microstructure of dried Angelica sinensis products. The results showed that microwave vacuum drying can significantly shorten the drying time and improved the drying efficiency. Six different mathematical models were investigated and the Midilli model was the best-fitted model for all samples (R² = 0.99903, Pearson’s r = 0.99952), and drying methods had various effects on different indexes and were confirmed by Pearson’s correlation analysis and principal component analysis. The optimal process parameters for microwave vacuum drying of Angelica sinensis were determined by entropy weight-coefficient of variation method as 45 °C, 4 mm, −0.70 kPa. Under this condition, well preserved of ferulic acid, senkyunolide I, senkyunolide H, ligustilide, total phenols and antioxidant activity, bright color (L* = 77.97 ± 1.89, ΔE = 6.77 ± 2.01), complete internal organizational structure and more regular cell arrangement were obtained in the samples. This study will provide a theoretical reference for the excavation of the potential value and the development of industrial processing of Angelica sinensis.