I am reading an image file using cloud vision. My input file is similar to a form. CV reads all the data close to accuracy of handwritten text (including numbers, alphabets). But it misses the order the text stored in the input image. My input image file contains multiple rows. Example. A row will be like A 86.43 B 232.63 C empty and so on multiple rows are there. Output become A B 232.63 C 86.43.... (A B C D... are printed text) And the values next to A B C D are handwritten. Each pair is a combination of Printed, Handwritten.
Feature.Type.TEXT_DETECTION
&& Feature.Type.DOCUMENT_TEXT_DETECTION
both tried.
Any other specific values to be set, in order to maintain the row order strictly ? Anyother ImageContext or Feature
need to be set? I am not set any ImageContext
as of now.
Input Image is
Output received was, Producer: Crop: Sample Type: Initial Sample Source Type: C Destination: pint: Bin No: SONY Station #: Specification: TF 3 CWRS 13.4 Represented Tonnes: 540 curan 2021 Lab Testing Fangber 13.4 14.5 33. Stwo 379 WG TWG PROT MST HVK MIDG&SWF FUS ERG % 3.2 Sev Sprtd 0.5 MIL TOT SPRTD 0.6 FRST SHR FAI.L NBR DKG FY1S4 Pre-Inspected comment (80): 2.5% sex sprld QC Comment (Facility): Please Check Grade, Falling Number. mst 14.7 Px 13.4 Lbs 58.2 Mil 3.3 Sevspts .2 o