qrcode is not only an information storage approach, but also a spatial localization sign. Compared to other spatial localization signs, qrcode is more accurate and more efficient to be detected. To achieve spatial l...
详细信息
qrcode is not only an information storage approach, but also a spatial localization sign. Compared to other spatial localization signs, qrcode is more accurate and more efficient to be detected. To achieve spatial localization by qrcode, detection is the essential procedure. Existing approaches perform well in regular light condition, but perform badly in complex light condition, because frame quality is extremely damaged by complex light condition. In the real world, complex light condition is very common but always unavoidable. Therefore, it is necessary and worthwhile to improve the under-complex-light qr code detection. In this paper, Vaccine-YOLOv10 (VCY) is proposed to enhance qr code detection capability in complex light condition. First, GhostConv and FasterC2f are introduced to replace the corresponding original modules of YOLOv10n. Second, Simulative Data Augment Algorithm (SDA) is proposed to simulate 5 types of complex light condition. Third, self-built Multi-Scene qrcode Dataset (MSQ) is augmented by SDA for VCY training. Compared to the baseline model YOLOv10n, VCY is improved on both lightweight and accuracy. Specifically, FPS reaches 150;GFLOPs reduces from 8.2 to 5.3;mAP50 increases from 0.877 to 0.905. code: https://***/AlexTraveling/Vaccine-YOLOv10.
qrcodes are extensively utilized in multiple domains including product management, security authentication, and intelligent Internet of Things. However, current qr code detection models can be hindered by complex bac...
详细信息
qrcodes are extensively utilized in multiple domains including product management, security authentication, and intelligent Internet of Things. However, current qr code detection models can be hindered by complex backgrounds, low real-time performance, and high resource consumption. To solve this problem, based on the You Only Look Once version 5 (YOLOv5) algorithm, we propose a lightweight real-time qr code detection algorithm called EBP-YOLOv5. This work introduces an Efficient Channel Attention mechanism in the backbone network to strengthen the network feature extraction capability. Bi-directional Feature Pyramid Network is utilized to replace the PANet module to enhance the network's ability to capture features along the path. Additionally, EBP-YOLOv5 undergoes lightweight operation by reducing the number of model parameters through sparse training. L1 regularization is incorporated into the loss function to prune the weights of the Batch Normalization layer. Finally, distillation learning is applied to enhance the accuracy of the network model. Extensive evaluations on a natural scene qrcode dataset demonstrate that EBP-YOLOv5 achieves superior performance. Compared to the original YOLOv5s algorithm, EBP-YOLOv5 enhances the average accuracy by 4.6% while reducing the model size to only 9MB. This significantly reduces the parameter count and computational load while maintaining a high average accuracy of 97%. Meanwhile, EBP-YOLOv5 outperforms other lightweight models in terms of parameter count, computational complexity, detection accuracy, and model size, and is more suitable for deployment on edge devices.
In this paper. we focused on how to improve the recognition rate of the qrcode in embedded systems of ordinary CMOS camera, and proposed an efficient pre-processing and detecting method for qrcode images with comple...
详细信息
ISBN:
(纸本)9783642324260
In this paper. we focused on how to improve the recognition rate of the qrcode in embedded systems of ordinary CMOS camera, and proposed an efficient pre-processing and detecting method for qrcode images with complex background or uneven illumination and an image binarization algorithm based on image blocks. Moreover. qrcode images which have geometric distortion or rotation can be fast corrected with our perspective transform matrix created by qrcode's finder patterns and alignment pattern. Experiments on WinCE embedded platform show that our image pre-processing and detecting methods can improve the recognition rate and accelerate the speed of the qrcode decoding.
Autonomous unmanned aerial vehicle(UAV)landing is a challenging task,especially on a moving platform in an unstructured *** such a scenario,successful UAV landing is mainly affected by poor UAV localization *** solve ...
详细信息
Autonomous unmanned aerial vehicle(UAV)landing is a challenging task,especially on a moving platform in an unstructured *** such a scenario,successful UAV landing is mainly affected by poor UAV localization *** solve this problem,we propose a coarse-to-fine visual autonomous UAV landing system based on an enhanced visual positioning *** landing platform is marked with a specially designed qrcode marker,which is developed to improve the landing accuracy when the UAV approaches the landing ***,we employ the you only look once framework to enhance the visual positioning accuracy,thereby promoting the landing platform detection when the UAV is flying far *** framework recognizes the qrcode and decodes the position of a UAV by the corner points of the qr ***,we use the Kalman filter to fuse the position data decoded from the qrcode with those from the inertia measurement unit ***,the position data are used for UAV landing with a developed hierarchical landing *** verify the effectiveness of the proposed system,we performed experiments in different environments under various light *** experimental results demonstrate that the proposed system can achieve UAV landing with high accuracy,strong adaptability,and *** addition,it can achieve accurate landing in different operating environments without external real-time kinematic global positioning system(RTK-GPS)signals,and the average landing error is 11.5 cm,which is similar to the landing error when using RTK-GPS signals as the ground truth.
One of the most important and widely used technique for automatic identification is the use of visual codes. Identifiers encoded in various symbols and patterns make electronic reading possible, that greatly helps and...
详细信息
One of the most important and widely used technique for automatic identification is the use of visual codes. Identifiers encoded in various symbols and patterns make electronic reading possible, that greatly helps and speeds up processing, e.g., at cashier lines, warehouse transactions, high speed processing places, production lines. The common codes designed using geometric patterns usually identify types or entities. However, patterns can be produced that, by their nature, are unique and thus can be used to validate originality or authenticity. In this paper, we focus on the automatic localization and recognition of a kind of natural feature identifier (NFI). We present an image processing algorithm that successfully locates NFI code region in an image taken by a mobile camera and extracts features of the NFI glitters that can be the basis for recognition. We also show preliminary experimental results.
暂无评论