Deep Learning-Based Apple Detection with Attention Module and Improved Loss Function in YOLO

• Published in: Remote sensing (Basel, Switzerland) Vol. 15; no. 6; p. 1516
• Main Authors: Sekharamantry, Praveen Kumar, Melgani, Farid, Malacarne, Jonni
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01-03-2023
• Subjects: Accuracy; Agriculture; Apple; apple detection; Apples; Atomizing; attention model; Automation; Background noise; Cameras; Comparative studies; Computer vision; Deep learning; Farms; Fruits; Harvesting; Horticulture; Innovations; Localization; Machine vision; New technology; Physical work; Remote sensing; Robots; Vision systems; YOLOv5; Italy
• ISSN: 2072-4292; 2072-4292
• DOI: 10.3390/rs15061516

Horticulture and agriculture are considered as the important pillars of any economy. Current technological advancements have led to the development of several new technologies which are useful in atomizing the agriculture process. Apple farming has a significant role in Italy’s agriculture domain where manual labor is widely employed for apple picking which can be replaced by automated robot mechanisms. However, these mechanisms are based on computer vision methods. These methods focus on detection, localization and tracking the apple fruits in given video frames. Later, appropriate actions can be taken to enhance the production and harvesting. Several techniques have been presented for apple detection, but complex background, noise and image blurriness are the major causes which can deteriorate the performance of the system. Thus, in this work, we present a deep learning-based scheme to detect apples which uses Yolov5 architecture in live apple farm images. We further improve the Yolov5 architecture by incorporating an adaptive pooling scheme and attribute augmentation model. This model detects the smaller objects and improves the feature quality to detect the apples in complex backgrounds. Moreover, a loss function is also incorporated to obtain the accurate bounding box which helps to maximize the detection accuracy. The comparative study shows that the proposed approach with the improved Yolov5 architecture achieves overall accuracy of 0.97, 0.99, and 0.98 in terms of precision, recall, and F1-score, respectively.