文献类型：期刊文献

英文题名：Autonomous Multiple Tramp Materials Detection in Raw Coal Using Single-Shot Feature Fusion Detector

作者：Li, Dongjun[1];Meng, Guoying[1];Sun, Zhiyuan[2];Xu, Lili[3]

第一作者：Li, Dongjun

通讯作者：Li, DJ[1]

机构：[1]China Univ Min & Technol, Sch Mech Elect & Informat Engn, Beijing 100083, Peoples R China;[2]Beijing Union Univ, Coll Appl Sci & Technol, Beijing 100012, Peoples R China;[3]Univ Sci & Technol Beijing, Community Serv & Management, Beijing 100083, Peoples R China

第一机构：China Univ Min & Technol, Sch Mech Elect & Informat Engn, Beijing 100083, Peoples R China

通讯机构：[1]corresponding author), China Univ Min & Technol, Sch Mech Elect & Informat Engn, Beijing 100083, Peoples R China.

年份：2022

卷号：12

期号：1

外文期刊名：APPLIED SCIENCES-BASEL

收录：;Scopus(收录号：2-s2.0-85121667105);WOS:【SCI-EXPANDED(收录号:WOS:000751158400001)】；

基金：Funding: This work was supported in part by the National Key Research and Development Program of China under Grant 2016YFC0600907.

语种：英文

外文关键词：coal; gangue; tramp materials; object detection; SSD; feature fusion

摘要：In the coal mining process, various types of tramp materials will be mixed into the raw coal, which will affect the quality of the coal and endanger the normal operation of the equipment. Automatic detection of tramp materials objects is an important process and basis for efficient coal sorting. However, previous research has focused on the detection of gangue, ignoring the detection of other types of tramp materials, especially small targets. Because the initial Single Shot MultiBox Detector (SSD) lacks the efficient use of feature maps, it is difficult to obtain stable results when detecting tramp materials objects. In this article, an object detection algorithm based on feature fusion and dense convolutional network is proposed, which is called tramp materials in raw coal single-shot detector (TMRC-SSD), to detect five types of tramp materials such as gangue, bolt, stick, iron sheet, and iron chain. In this algorithm, a modified DenseNet is first designed and a four-stage feature extractor is used to down-sample the feature map stably. After that, we use the dilation convolution and multi-branch structure to enrich the receptive field. Finally, in the feature fusion module, we designed cross-layer feature fusion and attention fusion modules to realize the semantic interaction of feature maps. The experiments show that the module we designed is effective. This method is better than the existing model. When the input image is 300 x 300 pixels, it can reach 96.12% MAP and 24FPS. Especially in the detection of small objects, the detection accuracy has increased by 4.1 to 95.57%. The experimental results show that this method can be applied to the actual detection of tramp materials objects in raw coal.