SSIG-SegPlate
收藏资源简介:
SSIG-SegPlate数据集由巴西联邦大学智能监控兴趣小组创建,专注于车牌自动识别系统中的字符分割任务。该数据集包含2000张巴西车牌的高分辨率图像,总计14000个字符,每个字符均配有精确的边界框注释。数据集的创建旨在为车牌字符分割技术提供一个标准化的评估平台,解决在不受控环境中车牌识别的挑战。数据集的应用领域包括交通监控、车辆访问控制和收费系统等,旨在通过精确的字符分割提高光学字符识别的准确性。
The SSIG-SegPlate dataset was developed by the Intelligent Surveillance Interest Group of the Federal University of Brazil, focusing on the character segmentation task in automatic license plate recognition systems. It contains 2000 high-resolution images of Brazilian license plates, totaling 14,000 individual characters, with precise bounding box annotations for each character. The dataset is designed to provide a standardized evaluation platform for license plate character segmentation technologies, addressing the challenges of license plate recognition in uncontrolled environments. Its application scenarios include traffic monitoring, vehicle access control, toll collection systems and other related fields, aiming to improve the accuracy of optical character recognition through precise character segmentation.

- 1Benchmark for License Plate Character Segmentation联邦大学 · 2016年



