make python script document layout analysis yolov5 tensorflow
Budget: €30 – €250 EUR
Create a python script for document layout analysis and extraction
Convert pdf documents in a folder to png images with 95dpi (documentname _pagenumber.png)
Run yolov5s (small) pretrained object detection model in tensorflow over the images
Yolov5s_name_name.pt weights model will be taken from a user specified directory
Extract the recognized objects inside the bounding box above a user defined confidence threshold (0.1) as images in native resolution as png
Save the images in a new directory within the document directory with the name: detectionconfidence_imagenumber_documentname_pagenumber.png
Create a richly annotated python script that is easy to read and run for a novice
Make variables for dpi values (pdf2image library)
Convert pdf documents in a folder to png images with 95dpi (documentname _pagenumber.png)
Run yolov5s (small) pretrained object detection model in tensorflow over the images
Yolov5s_name_name.pt weights model will be taken from a user specified directory
Extract the recognized objects inside the bounding box above a user defined confidence threshold (0.1) as images in native resolution as png
Save the images in a new directory within the document directory with the name: detectionconfidence_imagenumber_documentname_pagenumber.png
Create a richly annotated python script that is easy to read and run for a novice
Make variables for dpi values (pdf2image library)