Fast and accurate detection of kiwifruit in orchard using improved YOLOv3-tiny model
- PDF / 1,693,700 Bytes
- 23 Pages / 439.37 x 666.142 pts Page_size
- 23 Downloads / 196 Views
Fast and accurate detection of kiwifruit in orchard using improved YOLOv3‑tiny model Longsheng Fu1,3,4,5 · Yali Feng1,6 · Jingzhu Wu2 · Zhihao Liu1 · Fangfang Gao1 · Yaqoob Majeed5 · Ahmad Al‑Mallahi7 · Qin Zhang5 · Rui Li1 · Yongjie Cui1
© Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract 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 × 3 and 1 × 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 × 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 × 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. Keywords Data augmentation · Image detection · Deep learning · YOLOv3-tiny model · Convolutional kernel
* Longsheng Fu [email protected]; [email protected] Extended author information available on the last page of the article
13
Vol.:(0123456789)
Precision Agriculture
Introduction China is the largest country producing kiwifruits worldwide, with a yield of 2,390,287 t in 2016 from a cultivated area of 197,048 ha (UN FAO 2018). Within China, Shaanxi Province has the most significant production, accounting for approximately 70% and 33% of the Chinese and global productions, respectively (Hu et al. 2017). Harvesting kiwifruits in this area mainly depends on manual picking, which is labor-intensive (Fu et al. 2016), and introducing mechanical harvesting is needed. Kiwifruits are commercially grown on sturdy support structures such as T-bars and pergolas. The T-bar trellis is common in China because of its low cost (Lu et al. 2016). It consists of a 1.7-m h
Data Loading...
