Fragrant pears have been one of the most favorite types in pear industry. However, manual picking cannot fully meet the demands of large-scale production, due to the labor intensity. Mechanical picking has the potential to significantly reduce the labor demands for the high productivity. Among them, the visual recognition has been one of the most key elements among the various picking robots. The complex orchard environment with the variable lighting, weather conditions, foliage obstructions and fruit overlap has also posed the significant challenges on both conventional image processing and machine learning-based object detection, thereby impeding the accurate identification of fragrant pears. This study aims to improve the accuracy and detection speeds in such unstructured settings. An optimized YOLOv8n-based object detection was introduced to specifically identify the fragrant pear. A dataset was also comprised of 4 500 images of fragrant pears. A variety of conditions were considered using image collection and data augmentation, such as different lighting and foliage obstructions within orchard environments. The Min-Max normalization was employed to consider the fragrant pear harvesting. A comprehensive evaluation was conducted on five lightweight YOLO versions: YOLOv3-tiny, YOLOv5n, YOLO6n, YOLOv7-tiny, and YOLOv8n. Different weights were assigned to optimize the model selection, in terms of the mAP, precision, recall, inference time, parameters and model size. The network structure was streamlined for the object detection of fragrant pear with YOLOv8n as the baseline. Four key areas were focused to optimize the model. Initially, the network backbone was redesigned to promote the network inference speed. The impact of the C2f structure on speed was assessed to replace the certain redundant C2f components. Subsequently, an effective PConv module was integrated into the detection head of network, in order to better recognize the obscured images. A weight-sharing strategy was also coupled with the detection head parameters to reduce the parameters. Additionally, the combination of simSPPF and Inner IoU was further augmented the inference speed and bounding box regression performance. Comparative trials were then conducted on the object detection of fragrant pear. The results revealed that the superior comprehensive performance was achieved in the YOLOv8n. A higher weighted score of 83.4% was obtained after the Min-Max normalization, compared with the YOLOv3-tiny, YOLOv5n, YOLOv6n, and YOLOv7-tiny. Thus, the baseline network was selected for the subsequent research. Optimization experiments were carried out with C2f structures. The replacement of C2f was identified in the 15th and 2nd layers of the network with Conv and Conv_Res modules, in order to enhance the computational efficiency. The optimization scheme was determined for the backbone network. Ablation experiments were conducted to verify the efficacy of the various improved modules. The refined YOLOv8n was achieved the superior accuracy and detection speed on the fragrant pear dataset, with a 0.4 and 0.5 percentage point increase in the F_0.5 score and mean average precision, respectively, compared with the original YOLOv8n model. Detection speeds on GPU and CPU devices increased by 34.0% and 24.4%, respectively, indicating the rates of 99.4 and 15.3 frames per second, respectively. This high-precision and rapid detection can provide the valuable technical support to the real-time detection of fragrant pears in natural orchard environments. The improved model can also be expected to deploy into the fragrant pear picking robots.

Sea buckthorn is one type of the most favorite fruits rich in minerals. However, the short shelf life has limited food and medicinal applications. Among them, drying has been the most popular processing to prepare sea buckthorn. However, the existing hot-air drying cannot fully meet the large-scale production of sea buckthorn, particularly because of the long drying cycle, low quality, and high energy consumption. In this study, an infrared combined hot-air drying system was developed via temperature and humidity control. A five-layer material tray was also designed to increase the amount of sea buckthorn to be dried. An electric heating of infrared radiation was adopted with the carbon-fiber plate along a finned single-head electric heating tube. Radiation drying was then achieved in the sea buckthorn. An axial fan was used as the circulating fan of the dryer, and a steam generator was used as a humidifying device. A systematic optimization was made on the structure of the airflow distribution chamber in the drying device. A cylindrical spoiler and a square wind baffle were designed to simulate the velocity, temperature, and humidity fields inside the drying device using the numerical simulation program COMSOL. There was a great variation in the uneven temperature and humidity within the drying layers. The results demonstrate that the drying inhomogeneity was alleviated by altering the airflow between the drying layers or the size of the air outlet of each drying layer. A spoiler was added to increase the air velocity, in order to prevent uneven drying in various locations of the drying layer. The drying capacity was determined for the high-temperature and low-humidity drying medium in the free flow area. The maximum velocity deviation ratio between the five drying layers decreased to 0.88% in the modified structure of the drying chamber. Taking the sea buckthorn as the research subject, an experimental test was conducted on the infrared combined hot air-drying process using temperature and humidity control. The drying moisture ratio decreased exponentially with an increase in drying time. Drying temperatures shared a substantial influence on the drying duration of sea buckthorn. A systematic evaluation was utilized to acquire the highest comprehensive score at 75 ℃ under various drying temperatures using the hierarchical analysis. The weight of each drying characteristic and quality was then determined. Specifically, the maximum brightness, rehydration ratio, Vc retention rate, and total flavonoid content were obtained at 40 min under different conditions of medium humidity. The lowest drying energy consumption, the highest rehydration ratio, the highest Vc retention rate, and the total flavonoid content were obtained at 10% medium humidity. The highest overall score was obtained at 40% medium humidity under different drying durations. The optimal combination was achieved for sea buckthorn in the process of infrared combined hot air-drying using temperature and humidity control. The finding can provide a strong reference to improve the level of mechanization in the primary processing of sea buckthorn for the healthy and sustainable development of the food industry.

This study aims to reveal and compare the heat and mass transfer and the drying kinetic parameters of red jujube slices during hot air drying, infrared combined hot air drying, and infrared vacuum pulsation drying. A three-dimensional coupled model was established for the heat and mass transfer of jujube slices under the three drying modes, according to the control equations of Fick's law of diffusion, Antonin's equation and Beer Lambert's law. The reliability of the model was verified using experimental data. Subsequently, the utility of the model was tested by varying parameters, such as hot air temperatures, infrared radiation intensities, and vacuum pulsation ratios. Each drying was further analyzed to combine with the measured quality indexes, textural properties and microstructure. The actual geometry of jujube slices was simulated and solved using COMSOL Multiphysics 6.0. The result shows that: 1) The infrared hot air and infrared vacuum drying saved 46.43% and 41.07% of the drying time, compared with the hot air drying. The simulation was in better agreement with the measured values (the coefficients of determination R² for dry basis moisture content and temperature were 0.964, 0.959, 0.917 and 0.947, 0.922, 0.951, respectively). 2) The temperature field diagram showed that the effective heat of the inside of the material, and the center temperature of the material during infrared hot air and infrared vacuum drying increased by 11.33% and 5.59%, respectively, compared with the hot air drying, when drying for 20 min. The high energy density and penetrability of infrared radiation significantly improved the drying efficiency. 3) The simulation data showed that there was a significant effect of pressure variation on the drying kinetics in the infrared vacuum pulsation drying. The increasing vacuum time was favorable to the decrease of moisture. The water content and drying rate showed a step and peak distribution with the pressure pulsation. The sensitivity of the drying rate to the pressure change decreased with the decrease in the water content of the material. Further explanation was provided for the formation of porous structure in the red jujube slices during vacuum pulsation drying. 4) A comparison was made on the measured quality and textural properties of jujube slices with the simulated data. Segmented drying was proposed for the future direction in the numerical model of heat and mass transfer for fruit and vegetable drying. For example, the nutritional content and microstructure of products were coupled with the numerical model in the future. The nutritional, structural and rheological properties of materials can be expected to be quantified and visualized during drying. The numerical models of jujube slices were established and verified under three drying modes. The characteristics of each drying were obtained after simulation. It is of great significance to establish numerical models for fruit drying.

Drying is one of the major methods to extend the shelf life of fruits and vegetables. With the development of computer technology, computational fluid dynamics has been more and more widely used in the field of fruit and vegetable drying. This technique can simulate and predict the kinetic phenomena such as fluid flow, heat and mass transfer during the drying process of fruits and vegetables, and output the visualization results. Compared with traditional experiments, computational fluid dynamics has the advantages of energy saving, low cost, fast simulation speed and high flexibility. This article elaborates on the working principle of computational fluid dynamics in the field of fruit and vegetable drying, and reviews recent progress in the application of computational fluid dynamics to different fruit and vegetable drying methods as well as in the numerical modeling of coupled shrinkage. Finally, this review concludes with an outlook on future directions of computational fluid dynamics in the field of fruit and vegetable drying. It is our hope that this review will provide new insights and references for researchers.