BFP Net: Balanced Feature Pyramid Network for Small Apple Detection in Complex Orchard Environment

Meili Sun; Liancheng Xu; Xiude Chen; Ze Ji; Yuanjie Zheng; Weikuan Jia

doi:10.34133/2022/9892464

BFP Net: Balanced Feature Pyramid Network for Small Apple Detection in Complex Orchard Environment

Plant Phenomics. 2022 Sep 24:2022:9892464. doi: 10.34133/2022/9892464. eCollection 2022.

Authors

Meili Sun^{1

2}, Liancheng Xu¹, Xiude Chen³, Ze Ji⁴, Yuanjie Zheng¹, Weikuan Jia^{1

2}

Affiliations

¹ School of Information Science and Engineering, Shandong Normal University, Jinan, China.
² Key Laboratory of Facility Agriculture Measurement and Control Technology and Equipment of Machinery Industry, Zhenjiang 212013, China.
³ National Engineering Research Center for Apple, Shandong Agriculture University, Taian 271018, China.
⁴ School of Engineering, Cardiff University, Cardiff CF24 3AA, UK.

Abstract

Despite of significant achievements made in the detection of target fruits, small fruit detection remains a great challenge, especially for immature small green fruits with a few pixels. The closeness of color between the fruit skin and the background greatly increases the difficulty of locating small target fruits in the natural orchard environment. In this paper, we propose a balanced feature pyramid network (BFP Net) for small apple detection. This network can balance information mapped to small apples from two perspectives: multiple-scale fruits on the different layers of FPN and a characteristic of a new extended feature from the output of ResNet50 conv1. Specifically, we design a weight-like feature fusion architecture on the lateral connection and top-down structure to alleviate the small-scale information imbalance on the different layers of FPN. Moreover, a new extended layer from ResNet50 conv1 is embedded into the lowest layer of standard FPN, and a decoupled-aggregated module is devised on this new extended layer of FPN to complement spatial location information and relieve the problem of locating small apple. In addition, a feature Kullback-Leibler distillation loss is introduced to transfer favorable knowledge from the teacher model to the student model. Experimental results show that AP_S of our method reaches 47.0%, 42.2%, and 35.6% on the benchmark of the GreenApple, MinneApple, and Pascal VOC, respectively. Overall, our method is not only slightly better than some state-of-the-art methods but also has a good generalization performance.