2019年大创项目和2019年新苗人才计划项目,针对翻拍的图片进行检测与识别,主要使用lbp算法,前期准确率在80.08%左右,后不断扩大优化图片数据库,准确率提升到97.4%结题。
The 2019 Da Chuang project and the 2019 New Miao Talent Plan project mainly used the lbp algorithm for detection and recognition of remastered images, with an initial accuracy rate of around 80.08%. Later, they continuously expanded and optimized the image database, and the accuracy rate improved to 97.4%. (2021-03-03, Jupyter Notebook, 162342KB, 下载0次)