|
import os |
|
import gradio as gr |
|
from Plan.AiLLM import llm_recognition |
|
from Plan.pytesseractOCR import ocr_recognition |
|
from Preprocess.preprocessImg import ( |
|
preprocess_image001, preprocess_image002, preprocess_image003, |
|
preprocess_image004, preprocess_image005 |
|
) |
|
|
|
|
|
languages = os.popen('tesseract --list-langs').read().split('\n')[1:-1] |
|
|
|
|
|
def preprocess_and_ocr(image, valid_type, language): |
|
|
|
pre_img_001 = preprocess_image001(image) |
|
ocr_result_001 = ocr_recognition(pre_img_001, valid_type, language) |
|
|
|
pre_img_002 = preprocess_image002(image) |
|
ocr_result_002 = ocr_recognition(pre_img_002, valid_type, language) |
|
|
|
pre_img_003 = preprocess_image003(image) |
|
ocr_result_003 = ocr_recognition(pre_img_003, valid_type, language) |
|
|
|
pre_img_004 = preprocess_image004(image) |
|
ocr_result_004 = ocr_recognition(pre_img_004, valid_type, language) |
|
|
|
pre_img_005 = preprocess_image005(image) |
|
ocr_result_005 = ocr_recognition(pre_img_005, valid_type, language) |
|
|
|
return (pre_img_001, pre_img_002, pre_img_003, pre_img_004, pre_img_005, |
|
ocr_result_001, ocr_result_002, ocr_result_003, ocr_result_004, ocr_result_005) |
|
|
|
|
|
def preprocess_and_llm(image, valid_type, language): |
|
|
|
pre_img_001 = preprocess_image001(image) |
|
llm_result_001 = llm_recognition(pre_img_001, valid_type, language) |
|
|
|
pre_img_002 = preprocess_image002(image) |
|
llm_result_002 = llm_recognition(pre_img_002, valid_type, language) |
|
|
|
pre_img_003 = preprocess_image003(image) |
|
llm_result_003 = llm_recognition(pre_img_003, valid_type, language) |
|
|
|
pre_img_004 = preprocess_image004(image) |
|
llm_result_004 = llm_recognition(pre_img_004, valid_type, language) |
|
|
|
pre_img_005 = preprocess_image005(image) |
|
llm_result_005 = llm_recognition(pre_img_005, valid_type, language) |
|
|
|
return (pre_img_001, pre_img_002, pre_img_003, pre_img_004, pre_img_005, |
|
llm_result_001, llm_result_002, llm_result_003, llm_result_004, llm_result_005) |
|
|
|
|
|
with gr.Blocks() as demo: |
|
with gr.Row(): |
|
image_input = gr.Image(type="pil", label="上傳圖片") |
|
validation_type = gr.Dropdown(choices=["身分證正面", "身分證反面"], label="驗證類別") |
|
language_dropdown = gr.Dropdown(choices=languages, value="chi_tra", label="語言") |
|
|
|
with gr.Row(): |
|
ocr_button = gr.Button("使用 OCR") |
|
llm_button = gr.Button("使用 AI LLM") |
|
|
|
with gr.Row(): |
|
preprocess_output_001 = gr.Image(type="pil", label="預處理後的圖片-方案一") |
|
ocr_output_001 = gr.JSON(label="OCR-001-解析結果") |
|
llm_output_001 = gr.JSON(label="AiLLM-001-解析結果") |
|
with gr.Row(): |
|
preprocess_output_002 = gr.Image(type="pil", label="預處理後的圖片-方案二") |
|
ocr_output_002 = gr.JSON(label="OCR-002-解析結果") |
|
llm_output_002 = gr.JSON(label="AiLLM-002-解析結果") |
|
|
|
with gr.Row(): |
|
preprocess_output_003 = gr.Image(type="pil", label="預處理後的圖片-方案三") |
|
ocr_output_003 = gr.JSON(label="OCR-003-解析結果") |
|
llm_output_003 = gr.JSON(label="AiLLM-003-解析結果") |
|
with gr.Row(): |
|
preprocess_output_004 = gr.Image(type="pil", label="預處理後的圖片-方案四") |
|
ocr_output_004 = gr.JSON(label="OCR-004-解析結果") |
|
llm_output_004 = gr.JSON(label="AiLLM-004-解析結果") |
|
with gr.Row(): |
|
preprocess_output_005 = gr.Image(type="pil", label="預處理後的圖片-方案五") |
|
ocr_output_005 = gr.JSON(label="OCR-005-解析結果") |
|
llm_output_005 = gr.JSON(label="AiLLM-005-解析結果") |
|
|
|
ocr_button.click(preprocess_and_ocr, inputs=[image_input, validation_type, language_dropdown], |
|
outputs=[ |
|
preprocess_output_001, preprocess_output_002, preprocess_output_003, preprocess_output_004, |
|
preprocess_output_005, |
|
ocr_output_001, ocr_output_002, ocr_output_003, ocr_output_004, ocr_output_005 |
|
]) |
|
llm_button.click(preprocess_and_llm, inputs=[image_input, validation_type, language_dropdown], |
|
outputs=[ |
|
preprocess_output_001, preprocess_output_002, preprocess_output_003, preprocess_output_004, |
|
preprocess_output_005, |
|
llm_output_001, llm_output_002, llm_output_003, llm_output_004, llm_output_005 |
|
]) |
|
|
|
demo.launch(share=False) |
|
|
