Wang Chunchieh, Huang Chengwei, Zhu Xiaoming, Li Zhi, Zhao Liye
School of Instrument Science and Engineering, Southeast University, Nanjing, 210000, China.
Zhejiang Laboratory, Hangzhou, 310000, China.
Sci Rep. 2023 Dec 15;13(1):22512. doi: 10.1038/s41598-023-49543-x.
Intelligent signal processing in unmanned stores enhances operational efficiency, notably through automated SKUs (Stock Keeping Units) recognition, which expedites customer checkout. Distinguishing itself from generic detection algorithms, the retail product detection algorithm addresses challenges like densely arranged items, varying scales, large quantities, and product similarities. To mitigate these challenges, firstly we propose a novel boundary regression neural network architecture, which enhances the detection of bounding box in dense arrangement, minimizing computational costs and parameter sizes. Secondly, we propose a novel loss function for hierarchical detection, addressing imbalances in positive and negative samples. Thirdly, we enhance the conventional non-maximum suppression (NMS) with weighted non-maximum suppression (WNMS), tying NMS ranking scores to candidate box accuracy. Experimental results on SKU-110K and RPC datasets, two public available databases, show that the proposed SKUs recognition algorithm provides improved reliablity and efficiency over existing methods.
无人商店中的智能信号处理提高了运营效率，特别是通过自动识别库存保有单位（SKU），这加快了顾客结账速度。零售产品检测算法有别于一般的检测算法，它应对诸如物品密集排列、尺度变化、数量众多以及产品相似性等挑战。为了缓解这些挑战，首先我们提出了一种新颖的边界回归神经网络架构，它增强了在密集排列中对边界框的检测，将计算成本和参数规模降至最低。其次，我们提出了一种用于分层检测的新颖损失函数，解决正负样本的不平衡问题。第三，我们用加权非极大值抑制（WNMS）增强传统的非极大值抑制（NMS），将NMS排序分数与候选框准确性联系起来。在两个公开可用数据库SKU-110K和RPC数据集上的实验结果表明，所提出的SKU识别算法比现有方法具有更高的可靠性和效率。
