Small Object Detection via Precise Region-Based Fully Convolutional Networks
Authors:
- Dengyong ZhangDZDengyong Zhang
- Jiawei HuJHJiawei Hu
- Feng LiFLFeng Li
- Xiangling DingXDXiangling Ding
- Arun Kumar SangaiahArun Kumar Sangaiah
- Victor S. ShengVSVictor S. Sheng
Open Access
- 1 January 2021
- journal article
- research article
- Published by Tech Science Press in Computers, Materials & Continua
- Vol. 69 (2) , 1503-1517
Abstract
In the past several years, remarkable achievements have been made in the field of object detection. Although performance is generally improving, the accuracy of small object detection remains low compared with that of large object detection. In addition, localization misalignment issues are common for small objects, as seen in GoogLeNets and residual networks (ResNets). To address this problem, we propose an improved region-based fully convolutional network (R-FCN). The presented technique improves detection accuracy and eliminates localization misalignment by replacing position-sensitive region of interest (PS-RoI) pooling with position-sensitive precise region of interest (PS-Pr-RoI) pooling, which avoids coordinate quantization and directly calculates two-order integrals for position-sensitive score maps, thus preventing a loss of spatial precision. A validation experiment was conducted in which the Microsoft common objects in context (MS COCO) training dataset was oversampled. Results showed an accuracy improvement of for object detection tasks and an increase of for small objects.Keywords
This publication has 16 references indexed in Scilit:
- Aspect based sentiment analysis by a linguistically regularized CNN with gated mechanismJournal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology, 2019
- Intelligent Mobile Drone System Based on Real-Time Object DetectionJournal on Artificial Intelligence, 2019
- Mask R-CNNIEEE Transactions on Pattern Analysis and Machine Intelligence, 2018
- Generative Adversarial Networks: An OverviewIEEE Signal Processing Magazine, 2018
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksIEEE Transactions on Pattern Analysis and Machine Intelligence, 2016
- Spatial Pyramid Pooling in Deep Convolutional Networks for Visual RecognitionIEEE Transactions on Pattern Analysis and Machine Intelligence, 2015
- Selective Search for Object RecognitionInternational Journal of Computer Vision, 2013
- Distinctive Image Features from Scale-Invariant KeypointsInternational Journal of Computer Vision, 2004
- Robust Real-Time Face DetectionInternational Journal of Computer Vision, 2004
- Shape matching and object recognition using shape contextsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2002