期刊
PRECISION AGRICULTURE
卷 22, 期 3, 页码 754-776出版社
SPRINGER
DOI: 10.1007/s11119-020-09754-y
关键词
Data augmentation; Image detection; Deep learning; YOLOv3-tiny model; Convolutional kernel
资金
- Key Research and Development Program in Shaanxi Province of China [2018TSCXL-NY-05-04, 2019ZDLNY02-04]
- Fund of the Beijing Key Laboratory of Big Data Technology for Food Safety, Beijing Technology AMP
- Business University [BTBD2019KF03]
- International Scientific and Technological Cooperation Foundation of Northwest AF University [A213021803]
Automatic detection of kiwifruit in the orchard is challenging due to varying illumination and color similarity with the complex background. A deep YOLOv3-tiny model (DY3TNet) was developed for efficient and accurate detection, achieving high precision and fast processing speed, especially in images captured with flash.
Automatic detection of kiwifruit in the orchard is challenging because illumination varies through the day and night and because of color similarity between kiwifruit and the complex background of leaves, branches and stems. Also, kiwifruits grow in clusters, which may result in having occluded and touching fruits. A fast and accurate object detection algorithm was developed to automatically detect kiwifruits in the orchard by improving the YOLOv3-tiny model. Based on the characteristics of kiwifruit images, two convolutional kernels of 3 x 3 and 1 x 1 were added to the fifth and sixth convolution layers of the YOLOv3-tiny model, respectively, to develop a deep YOLOv3-tiny (DY3TNet) model. It takes multiple 1 x 1 convolutional layers in intermediate layers of the network to reduce the computational complexity. Testing images captured from day and night and comparing with other deep learning models, namely, Faster R-CNN with ZFNet, Faster R-CNN with VGG16, YOLOv2 and YOLOv3-tiny, the DY3TNet model achieved the highest average precision of 0.9005 with the smallest data weight of 27 MB. Furthermore, it took only 34 ms on average to process an image of a resolution of 2352 x 1568 pixels. The DY3TNet model, along with the YOLOv3-tiny model, showed better performance on images captured with flash than those without. Moreover, the experiments indicated that the image augmentation process could improve the detection performance, and a simple lighting arrangement could improve the success rate of detection in the orchard. The experimental results demonstrated that the improved DY3TNet model is small and efficient and that it would increase the applicability of real-time kiwifruit detection in the orchard even when small hardware devices are used.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据