Citrus psyllid is recognized as the main vector of HLB thus, development of accurate detection technology for this pathogen will contribute to controlling and preventing the transmission of HLB. Furthermore, current methods require significant professional knowledge and skill as well as technology that is difficult to apply to citrus orchards. Biochemical analyses are complex and time consuming. Thus, it is difficult to achieve sufficient accuracy from the subjective judgment of inspectors. While this is a simple and easy method which does not require equipment, the environment of a citrus orchard is complex. Detection of HLB typically includes field diagnosis and laboratory biochemical analysis, with field diagnosis relying on manual diagnosis and identification. The disease can cause huge economic losses to the global citrus industry. Other symptoms of HLB include a decrease in fruit production, premature fruit drop, and twig dieback. Fruits from infected trees have a bitter taste and may have a yellow-green or greenish-brown color. Symptoms of HLB in citrus fruit trees include yellowing and mottling of the leaves, reduced leaf size, and stunted growth. These results confirm that the YOLOv5s-BC algorithm has good generalization ability in the natural context of citrus orchards, and it offers a new approach for the control of citrus psyllid.Ĭitrus Huanglongbing (HLB) is a devastating disease which affects citrus plants and is caused by infestation of the phloem by Gram-negative bacteria. The accuracy and recall rates are also increased by 1.31% and 4.22%, respectively. Experimental results based on a standard sample database show the recognition accuracy (intersection over union (IoU) = 0.5) of the YOLOv5s-BC algorithm for citrus psyllid to be 93.43%, 2.41% higher than that of traditional YOLOv5s. At the same time, the BottleneckCSP module in the neck network is improved, and extraction of multiple features of recognition targets is improved by the addition of a normalization layer and SiLU activation function. Based on YOLOv5s, our algorithm incorporates an SE-Net channel attention module into the Backbone network and improves the detection of small targets by guiding the algorithm to the channel characteristics of small-target information. By integrating the attention mechanism and optimizing the key module of BottleneckCSP, YOLOv5s-BC, we have created an accurate detection algorithm for small targets. The present paper describes the construction of a standard sample database of citrus psyllid in multi-focal lengths and out-of-focus states in the natural environment. The small size of this pest, difficulties in data acquisition, and the lack of target detection algorithms suitable for complex occlusion environments inhibit detection of the pest. Citrus psyllid is the main vector of Huanglongbing, and as such, it is responsible for huge economic losses across the citrus industry.
0 Comments
Leave a Reply. |