#AO file tools
Define a schema for storing image statistics, and some tools for creating and accessing them.
Collect these features
| Attribute | Description | data source | method | data attribute |
|---|---|---|---|---|
| storage_file | string of file path (root) | import | SQL update | key of file |
| format | ('jpg', 'png', 'single_image_tiff', 'jp2', 'raw') | PIL | PIL Image.open() | .format |
| mode | ('1', 'L', 'RGB', 'RGBA', 'CMYK', 'P', 'OTHER') | PIL | PIL Image.open() | .format |
| tiff_compression | ('raw', 'tiff_ccitt', 'group3', 'group4', 'tiff_lzw', 'tiff_jpeg', 'jpeg', 'tiff_adobe_deflate', 'lzma', 'other') |
PIL | Image.open() | .info.get('compression', 'unknown') |
| width | width of the bitmap (not taking a potential exif rotation into account)' | PIL | Image.open() | .img.size[0] |
| height | height of the bitmap (not taking a potential exif rotation into account)' | PIL | Image.open() | .img.size[1] |
| quality | relevant only for jpg, png and single_image_tiff encoded as jpg: quality of encoding. JPEG is represented between 0 and 100. For PNG this column encodes the compression between 0 and 9.' |
PIL | Image.open | .info.get('quality', 'unknown') |
| bps | 'bits per sample' aka resolution | PIL | Image.open() | .info.get('dpi', 'unknown') |
| recorded_date | COMMENT 'the timestamp recorded in the exif metadata' | PIL | Image.open() | img._getexif() |
Collect these features
`number_of_pages` smallint UNSIGNED COMMENT 'the number of pages',
`median_nb_chr_per_page` smallint UNSIGNED COMMENT 'the average number of characters in a page',
`median_nb_images_per_page` smallint UNSIGNED COMMENT 'the average number of images per page',
`recorded_date` timestamp COMMENT 'the timestamp recorded in the exif metadata'| Attribute | Description | data source | method | data attribute |
|---|---|---|---|---|
| number_of_pages | the number of pages | pypdf.PdfFileReader | .numPages | |
| median_nb_chr_per_page | the average number of characters in a page | pypdf.PdfFileReader | .numPages | |
| median_nb_images_per_page | the average number of images per page | pypdf.PdfFileReader | .numPages | |
| recorded_date | the timestamp recorded in the meetadata | pypdf.PdfFileReader | .document_info | ["/CreationDate"] |