A two-stage algorithm based on deep learning for the detection and recognition of can bottom spray codes and numbers is proposed to address the problems of small character areas and fast production line speeds in can ...A two-stage algorithm based on deep learning for the detection and recognition of can bottom spray codes and numbers is proposed to address the problems of small character areas and fast production line speeds in can bottom spray code number recognition.In the coding number detection stage,Differentiable Binarization Network is used as the backbone network,combined with the Attention and Dilation Convolutions Path Aggregation Network feature fusion structure to enhance the model detection effect.In terms of text recognition,using the Scene Visual Text Recognition coding number recognition network for end-to-end training can alleviate the problem of coding recognition errors caused by image color distortion due to variations in lighting and background noise.In addition,model pruning and quantization are used to reduce the number ofmodel parameters to meet deployment requirements in resource-constrained environments.A comparative experiment was conducted using the dataset of tank bottom spray code numbers collected on-site,and a transfer experiment was conducted using the dataset of packaging box production date.The experimental results show that the algorithm proposed in this study can effectively locate the coding of cans at different positions on the roller conveyor,and can accurately identify the coding numbers at high production line speeds.The Hmean value of the coding number detection is 97.32%,and the accuracy of the coding number recognition is 98.21%.This verifies that the algorithm proposed in this paper has high accuracy in coding number detection and recognition.展开更多
In unmanned aerial vehicle(UAV)-based scenarios,sensing-aided integrated data and energy networking(IDEN)systems can significantly mitigate non-line-of-sight(NLoS)propagation,thereby enhancing sensing accuracy..Howeve...In unmanned aerial vehicle(UAV)-based scenarios,sensing-aided integrated data and energy networking(IDEN)systems can significantly mitigate non-line-of-sight(NLoS)propagation,thereby enhancing sensing accuracy..However,the rapid channel variations induced by UAV mobility pose a challenge for traditional polar code construction methods,making it difficult to satisfy the stringent requirements of IDEN systems.To address this challenge,we propose a neural network(NN)-based sensing-aided IDEN framework.This system leverages sensing information to assist polar code construction while satisfying energy constraints.Furthermore,it incorporates neural networks to optimize the performance of polar codes in dynamic environments.Specifically,a sensing-aided binarized neural network(BNN)-based polar encoder is proposed for both lowlatency and high-reliability requirements,and a deep neural network(DNN)-based polar decoder is applied to match the encoder.Moreover,the corresponding training method is proposed,which focuses on the initialization design of the NNs.The simulation results show that the NN-based sensing-aided polar encoding scheme outperforms the conventional counterparts in terms of IDEN for both low-latency and high-reliability requirements.展开更多
文摘A two-stage algorithm based on deep learning for the detection and recognition of can bottom spray codes and numbers is proposed to address the problems of small character areas and fast production line speeds in can bottom spray code number recognition.In the coding number detection stage,Differentiable Binarization Network is used as the backbone network,combined with the Attention and Dilation Convolutions Path Aggregation Network feature fusion structure to enhance the model detection effect.In terms of text recognition,using the Scene Visual Text Recognition coding number recognition network for end-to-end training can alleviate the problem of coding recognition errors caused by image color distortion due to variations in lighting and background noise.In addition,model pruning and quantization are used to reduce the number ofmodel parameters to meet deployment requirements in resource-constrained environments.A comparative experiment was conducted using the dataset of tank bottom spray code numbers collected on-site,and a transfer experiment was conducted using the dataset of packaging box production date.The experimental results show that the algorithm proposed in this study can effectively locate the coding of cans at different positions on the roller conveyor,and can accurately identify the coding numbers at high production line speeds.The Hmean value of the coding number detection is 97.32%,and the accuracy of the coding number recognition is 98.21%.This verifies that the algorithm proposed in this paper has high accuracy in coding number detection and recognition.
基金supported in part by the Natural Science Foundation of China under Grants 62132004,62531008,and 62301122in part by Jiangsu Major Project on Fundamental Research under Grant BK20243059+3 种基金in part by Gusu Innovation Project under Grant ZXL2024360in part by the High-Tech District of Suzhou City under Grant RC2025001in part by Quzhou Government under Grants 2024D007,and 2023D005in part by the Xiaomi Young Scholar Award,in part by Nanjing University-China Mobile Communications Group Co.,Ltd.Joint Institute.
文摘In unmanned aerial vehicle(UAV)-based scenarios,sensing-aided integrated data and energy networking(IDEN)systems can significantly mitigate non-line-of-sight(NLoS)propagation,thereby enhancing sensing accuracy..However,the rapid channel variations induced by UAV mobility pose a challenge for traditional polar code construction methods,making it difficult to satisfy the stringent requirements of IDEN systems.To address this challenge,we propose a neural network(NN)-based sensing-aided IDEN framework.This system leverages sensing information to assist polar code construction while satisfying energy constraints.Furthermore,it incorporates neural networks to optimize the performance of polar codes in dynamic environments.Specifically,a sensing-aided binarized neural network(BNN)-based polar encoder is proposed for both lowlatency and high-reliability requirements,and a deep neural network(DNN)-based polar decoder is applied to match the encoder.Moreover,the corresponding training method is proposed,which focuses on the initialization design of the NNs.The simulation results show that the NN-based sensing-aided polar encoding scheme outperforms the conventional counterparts in terms of IDEN for both low-latency and high-reliability requirements.
