Litcius/Paper detail

YOLO-MSA: A Multiscale Stereoscopic Attention Network for Empty-Dish Recycling Robots

Xuebin Yue, Lin Meng

2023IEEE Transactions on Instrumentation and Measurement21 citationsDOI

Abstract

As the global population ages and the labor force shrinks, using Artificial Intelligence (AI) technology to promote labor productivity growth has become a hot topic. The emergence of Empty-Dish Recycling Robots has effectively alleviated the impact of the decline in labor productivity. This paper proposes a Multi-scale Stereoscopic Attention (MSA) network YOLO-MSA to detect postprandial dishes for Empty-Dish Recycling Robots. First, the standard convolution is replaced with a Res2Net module, which improves the multi-scale expressiveness of the network at a finer-grained level. Second, we adopt a Res2Net with different dilation rates and a novel stereoscopic attention mechanism to propose an MSA module, which is used for coarse-grained multi-scale expression. Thirdly, for multi-scale feature learning in the dimensionality reduction process, a Dimension Reduction Spatial Pyramid Pooling (DRSPP) is proposed to fuse feature maps of different scales. Extensive experiments demonstrate the effectiveness of the proposed MSA module for multi-scale feature learning. Furthermore, YOLO-MSA has achieved 98.47% mean Average Precision ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">mAP</i> ) on Dish-21, a dataset of the postprandial dishes, which is much higher than other state-of-the-art models, and has achieved an inference speed of 33.93 frames per second ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">FPS</i> ), which meets the needs of real-time detection of the postprandial dish for the Empty-Dish Recycling Robot. Test results on other public datasets show that the proposed YOLO-MSA has a better generalization ability. In summary, YOLO-MSA exhibits satisfactory multi-scale feature expression ability, demonstrates effectiveness and robustness in postprandial dish detection, and has far-reaching significance for the development of Empty-Dish Recycling Robots.

Topics & Concepts

Artificial intelligenceStereoscopyScale (ratio)RobotComputer scienceFeature (linguistics)Computer visionCartographyGeographyLinguisticsPhilosophyAdvanced Neural Network ApplicationsIndustrial Vision Systems and Defect DetectionFace recognition and analysis