You should try table extraction from Amazon Textract. They go a step beyond OCR and try to detect tabular regions. Table detection is immensely complicated ...
確定! 回上一頁