车牌自动识别库：OpenALPR
jopen
8年前
OpenALPR 是一个采用C++编写的开源车牌自动识别库，并绑定了 C#、Java、Node.js 和 Python支持。此库通过分析图像和视频流来识别车牌。输出的结果是车牌字符的文本类型。
user@linux:~/openalpr$ alpr ./samplecar.png plate0: top 10 results -- Processing Time = 58.1879ms. - PE3R2X confidence: 88.9371 - PE32X confidence: 78.1385 - PE3R2 confidence: 77.5444 - PE3R2Y confidence: 76.1448 - P63R2X confidence: 72.9016 - FE3R2X confidence: 72.1147 - PE32 confidence: 66.7458 - PE32Y confidence: 65.3462 - P632X confidence: 62.1031 - P63R2 confidence: 61.5089
Detailed command line usage:
user@linux:~/openalpr$ alpr --help USAGE: alpr [-c <country_code>] [--config <config_file>] [-n <topN>] [--seek <integer_ms>] [-p <pattern code>] [--clock] [-d] [-j] [--] [--version] [-h] <image_file_path> Where: -c <country_code>, --country <country_code> Country code to identify (either us for USA or eu for Europe). Default=us --config <config_file> Path to the openalpr.conf file -n <topN>, --topn <topN> Max number of possible plate numbers to return. Default=10 --seek <integer_ms> Seek to the specified millisecond in a video file. Default=0 -p <pattern code>, --pattern <pattern code> Attempt to match the plate number against a plate pattern (e.g., md for Maryland, ca for California) --clock Measure/print the total time to process image and all plates. Default=off -d, --detect_region Attempt to detect the region of the plate image. [Experimental] Default=off -j, --json Output recognition results in JSON format. Default=off --, --ignore_rest Ignores the rest of the labeled arguments following this flag. --version Displays version information and exits. -h, --help Displays usage information and exits. <image_file_path> Image containing license plates OpenAlpr Command Line Utility
