ETHSeg: An Amodel Instance Segmentation Network and a Real-world Dataset for X-Ray Waste Inspection
Lingteng Qiu1,2,3
Zhangyang Xiong1,2
Xuhao Wang2,3
KenKun Liu2
Yihan Li2
Guanying Chen1,2
Xiaoguang Han1,2*
Shuguang Cui1,2,3
*Corresponding email: hanxiaoguang@cuhk.edu.cn
1The Future Network of Intelligence Institute, CUHK-Shenzhen
2School of Science and Engineering, CUHK-Shenzhen
3Shenzhen Research Institute of Big Data
Introduction
Waste inspection for packaged waste is an important step in the pipeline of waste disposal. Previous methods either rely on manual visual checking or RGB image-based inspection algorithm, requiring costly preparation procedures (e.g., open the bag and spread the waste items). Moreover, occluded items are very likely to be left out. Inspired by the fact that X-ray has a strong penetrating power to see through the bag and overlapping objects, we propose to perform waste inspection efficiently using X-ray images without the need to open the bag. We introduce a novel problem of instance-level waste segmentation in X-ray image for intelligent waste inspection, and contribute a real dataset consisting of 5,038 X-ray images (totally 30,881 waste items) with high-quality annotations (i.e., waste categories, object boxes, and instance-level masks) as a benchmark for this problem. As existing segmentation methods are mainly designed for natural images and cannot take advantage of the characteristics of X-ray waste images (e.g., heavy occlusions and penetration effect), we propose a new instance segmentation method to explicitly take these image characteristics into account. Specifically, our method adopts an easy-to-hard disassembling strategy to use high confidence predictions to guide the segmentation of highly overlapped objects, and a global structure guidance module to better capture the complex contour information caused by the penetration effect. Extensive experiments demonstrate the effectiveness of the proposed method.
Download
If you use ETHSeg in your work, please consider citing our paper!
@article{
Dataset
According to current waste disposal scenarios, we classify the domestic wastes into four general types and twelve categories: Recyclable (PlasticBottle, Can, Carton, GlassBottle, Stick, and Tableware), Foodwaste (FoodWaste), Residual (HeatingPad, Desiccant, and MealBox), and Hazardous (Battery and Bulb). In total, our WIXray contains 5, 038 X-ray images and 30, 845 waste instances covering 12 common waste categories. The following table summarizes the statistics of the introduced dataset. Unlike existing X-ray datasets for security inspection which only annotate a few forbidden objects, we densely labeled the common waste items in the picture.
Methodology
Quantitative Results
Qualitative Results
