基于CBAM-YOLOv8的温室番茄果实识别

打开文本图片集

Greenhouse tomato fruit recognition based on CBAM-YOLOv8

Abstract：Efficientandaccuratedetectionoftomatofruitsisthekeytoachievinginteligentharvesting.Toaddresthe challenge of balancingaccuracyandspeed inexistingdeep learning object dectection models，this study proposed animproved YOLOv8 model.This model incroporated the CBAM atention module during the YOLOv8 feature extraction stage，whichdynamicallyadjustedfeatureweights toectivelysuppressnoiseandirrelevant information，andproved the accuracy of model detection. The experiments showed that CBAM-YOLOv8 performed wellin tomato detection， with accuracy，recall，and average precision reaching 91% ， 78% ，and 90% ，respectively.Compared with SSD，Faster RCNN，and original YOLOv8，the performance had significantly improved.This modelefectively reducedtheratesof false positives and false negatives.Interms of predictiontime，YOLOv8 has the shortest comprehensive time consumption.Incontrast，the prediction timerequired byCBAM-YOLOv8 model has increased，andthe inference speed is slower， therebyincreasingthecomputationalcost.Therefore，inpracticalapplications，abalanceneeds tobemadebetweenaccuracyand speed，Inconclusion，CBAM-YOLOv8 providesaneffectivesolutionforreal-time monitoring，yieldestimation， and efficient harvesting of tomato fruits.

Keywords： Tomato recognition; Target detection; Attention mechanism; CBAM-YOLOv8

在众多广泛种植的蔬菜作物中，番茄具有不可忽视的重要营养和经济价值。（剩余13041字）