Share Email Print

Proceedings Paper

Multi-size object detection assisting fault diagnosis of power systems based on improved cascaded faster R-CNNs
Author(s): Xiuhong Zhu; Lanfang Kong; Guoyou Wang; Zhaoyang Hu; Shanjun Li
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Recently, object detection has been widely used in power systems to assist fault diagnosis of transmission lines. However, it is still faced with great challenges due to multi-size targets existing in a single inspection image. Current state-of-art object detection pipelines, like Faster R-CNN, perform well on large objects with low resolution, but usually fail to detect small objects due to low resolution and poor representation. Many existing object detectors for this problem typically exploit feature pyramids, multi-scale image inputs, etc., which can attain high accuracy but is computation and memory consuming. In this paper, we propose an improved cascaded Faster R-CNNs framework that reduces the computational cost while maintaining high detection accuracy to cope with multi-size object detection in high-resolution inspection images, where the first-stage Faster R-CNN is used to detect large objects while the second one detects small objects relative to large objects. We further merge the first-stage and the second into a single network by sharing convolutional features–using the semantic context between multi-size targets, the first stage tells the second where to look. For the "tell" step, we just map the bounding box coordinates of large objects detected in the first stage to the VGG16 network, crop the corresponding feature maps and feed them to the following second stage. Experiments on the test datasets demonstrate that our method achieves a higher detection mAP of 87.6% at 5FPS on an NVidia Titan X compared with the one-stage Faster R-CNN.

Paper Details

Date Published: 9 August 2018
PDF: 10 pages
Proc. SPIE 10806, Tenth International Conference on Digital Image Processing (ICDIP 2018), 108061A (9 August 2018); doi: 10.1117/12.2503064
Show Author Affiliations
Xiuhong Zhu, Huazhong Univ. of Science and Technology (China)
Lanfang Kong, Huazhong Univ. of Science and Technology (China)
Guoyou Wang, Huazhong Univ. of Science and Technology (China)
Zhaoyang Hu, Huazhong Univ. of Science and Technology (China)
Shanjun Li, Huazhong Univ. of Science and Technology (China)

Published in SPIE Proceedings Vol. 10806:
Tenth International Conference on Digital Image Processing (ICDIP 2018)
Xudong Jiang; Jenq-Neng Hwang, Editor(s)

© SPIE. Terms of Use
Back to Top