shihaam/recipt-ocr-api
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

added status, better date and time and some processing

Browse Source
This commit is contained in:
Shihaam Abdul Rahman 2024-08-03 01:50:43 +05:00
parent 8979b754dd
commit 9e7bd54b1f
Signed by: shihaam
GPG Key ID: 6DA2E87EBC227636
1 changed files with 9 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
ocr.sh
View File

@ -1,18 +1,20 @@
#!/bin/bash #!/bin/bash
INPUT_IMAGE=$1 INPUT_IMAGE=$1
raw_text=$(tesseract $INPUT_IMAGE stdout) raw_text=$(tesseract $INPUT_IMAGE stdout --psm 6)
referece=$(echo "$raw_text" | grep -o 'BLAZ[0-9]*') referece=$(echo "$raw_text" | grep -o 'BLAZ[0-9]*')
to_number=$(echo "$raw_text" | grep -oE '\b9[0-9]{16}\b|\b7[0-9]{12}\b') to_number=$(echo "$raw_text" | grep -oE '\b9[0-9]{16}\b|\b7[0-9]{12}\b')
date=$(echo "$raw_text" | grep -oE '\b[0-3][0-9]/[0-1][0-9]/[0-9]{4}\b') #date=$(echo "$raw_text" | grep -oE '\b[0-3][0-9]/[0-1][0-9]/[0-9]{4}\b')
time=$(echo "$raw_text" | grep -oE '\b[0-2][0-9]:[0-5][0-9]\b') #time=$(echo "$raw_text" | grep -oE '\b[0-2][0-9]:[0-5][0-9]\b')
date=$(echo "$raw_text" | grep -oE '[0-3][0-9]/[0-1][0-9]/[0-9]{4}')
time=$(echo "$raw_text" | grep -oE '[0-2][0-9]:[0-5][0-9]')
currency=$(echo "$raw_text" | grep -oP '(?<=Amount\s)[A-Z]{3}') currency=$(echo "$raw_text" | grep -oP '(?<=Amount\s)[A-Z]{3}')
amount=$(echo "$raw_text" | grep -oP '(?<=Amount\s[A-Z]{3}\s)[0-9]{1,3}(?:,[0-9]{3})*(?:\.[0-9]{2})') amount=$(echo "$raw_text" | grep -oP '(?<=Amount\s[A-Z]{3}\s)[0-9]{1,3}(?:,[0-9]{3})*(?:\.[0-9]{2})')
status=$(echo "$raw_text" | grep -oP '(?<=Status\s)[A-Z]+')
json_output=$(cat <<EOF json=$(cat <<EOF
{ {
"Amount": "$amount", "Amount": "$amount",
"Currency": "$currency", "Currency": "$currency",
@ -30,4 +32,5 @@ json_output=$(cat <<EOF
EOF EOF
) )
echo $json_output echo $json | jq
echo $raw_text