Enhancing citrus fruit yield investigations through flight height optimization with UAV imaging.

Citrus fruit yield is essential for market stability, as it allows businesses to plan for production and distribution. However, yield estimation is a complex and time-consuming process that often requires a large number of field samples to ensure representativeness. To address this challenge, we investigated the optimal altitude for unmanned aerial vehicle (UAV) imaging to estimate the yield of Citrus unshiu fruit. We captured images from five different altitudes (30 m, 50 m, 70 m, 90 m, and 110 m), and determined that a resolution of approximately 5 pixels/cm is necessary for reliable estimation of fruit size based on the average diameter of C. unshiu fruit (46.7 mm). Additionally, we found that histogram equalization of the images improved fruit count estimation compared to using untreated images. At the images from 30 m height, the normal image estimates fruit numbers as 73, 55, and 88. However, the histogram equalized image estimates 88, 71, 105. The actual number of fruits is 124, 88, and 141. Using a Vegetation Index such as IPCA showed a similar estimation value to histogram equalization, but I1 estimation represents a gap to actual yields. Our results provide a valuable database for future UAV field investigations of citrus fruit yield. Using flying platforms like UAVs can provide a step towards adopting this sort of model spanning ever greater regions at a cheap cost, with this system generating accurate results in this manner.

Real-Time Detection and Classification of Scirtothrips dorsalis on Fruit Crops with Smartphone-Based Deep Learning System: Preliminary Results.

This study proposes a deep-learning-based system for detecting and classifying Scirtothrips dorsalis Hood, a highly invasive insect pest that causes significant economic losses to fruit crops worldwide. The system uses yellow sticky traps and a deep learning model to detect the presence of thrips in real time, allowing farmers to take prompt action to prevent the spread of the pest. To achieve this, several deep learning models are evaluated, including YOLOv5, Faster R-CNN, SSD MobileNetV2, and EfficientDet-D0. EfficientDet-D0 was integrated into the proposed smartphone application for mobility and usage in the absence of Internet coverage because of its smaller model size, fast inference time, and reasonable performance on the relevant dataset. This model was tested on two datasets, in which thrips and non-thrips insects were captured under different lighting conditions. The system installation took up 13.5 MB of the device's internal memory and achieved an inference time of 76 ms with an accuracy of 93.3%. Additionally, this study investigated the impact of lighting conditions on the performance of the model, which led to the development of a transmittance lighting setup to improve the accuracy of the detection system. The proposed system is a cost-effective and efficient alternative to traditional detection methods and provides significant benefits to fruit farmers and the related ecosystem.

Sensitive and Rapid Detection of Citrus Scab Using an RPA-CRISPR/Cas12a System Combined with a Lateral Flow Assay.

Citrus is the most extensively produced fruit tree crop in the world and is grown in over 130 countries. Fungal diseases in citrus can cause significant losses in yield and quality. An accurate diagnosis is critical for determining the best management practices and preventing future losses. In this study, a Recombinase polymerase amplification (RPA)-clustered regularly interspaced short palindromic repeats (CRISPR)/associated (Cas) system was established with the integration of a lateral flow assay (LFA) readout system for diagnosis of citrus scab. This detection can be completed within 1 h, is highly sensitive and prevents cross-reactions with other common fungal citrus diseases. Furthermore, the detection system is compatible with crude DNA extracted from infected plant tissue. This RPA-CRISPR/Cas12a-LFA system provides a sensitive, rapid, and cost-effective method with promising and significant practical value for point-of-care diagnosis of citrus scab. To our knowledge, this is the first report to establish an RPA- and CRISPR-based method with LFA for fungal diseases in plants.