Update Preprocess

app.py

@@ -3,6 +3,7 @@ import gradio as gr
 import cv2
 import pytesseract
 from fastapi import FastAPI
+import numpy as np
 
 #pytesseract.pytesseract.tesseract_cmd = r'./Tesseract-OCR/tesseract.exe'
 
@@ -11,8 +12,34 @@ print(os.popen(f'cat /etc/issue').read())
 print(os.popen(f'apt search tesseract').read())
 
 
+def PreprocessIMG(image):
+    
+    # Convert to grayscale
+    image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
+    # Apply Canny edge detection
+    image = cv2.GaussianBlur(image,(3,3),0)
+    edges = cv2.Canny(image, 90, 120, apertureSize=3)
+    # Apply Hough line transform to detect lines
+    lines = cv2.HoughLines(edges, 1, np.pi/180, threshold=300)
+
+    # Remove lines from the image
+    for rho, theta in lines[:, 0]:
+        a = np.cos(theta)
+        b = np.sin(theta)
+        x0 = a * rho
+        y0 = b * rho
+        x1 = int(x0 + 1000 * (-b))
+        y1 = int(y0 + 1000 * (a))
+        x2 = int(x0 - 1000 * (-b))
+        y2 = int(y0 - 1000 * (a))
+        cv2.line(image, (x1, y1), (x2, y2), (255, 255, 255), 2)
+
+    return image
+
 
-def TextLineBox(img):
+
+
+def TextLineBox(img, engine):
 
     class Lines:
         def __init__(self,x,y,w,h,text):
@@ -25,11 +52,10 @@ def TextLineBox(img):
     lineboxes = []
 
     #read image
-    img = cv2.GaussianBlur(img,(1,1),0)
-
+    img = PreprocessIMG(img)
 
     ### Cofig
-    configname = r' --oem 3 --psm ' + str(4) + ' -l eng'
+    configname = r' --oem 3 --psm ' + str(engine) + ' -l eng'
 
     #### Text for testing
     texttest = pytesseract.image_to_string(img ,config=configname)
@@ -90,6 +116,7 @@ with gr.Blocks (theme="ParityError/Anime"  , css="#SUBMIT {background-color: #cd
     with gr.Row():
         with gr.Column():
             input = gr.Image()
+            engine_input = gr.Text(label="Engine Mode Number")
             text_output = gr.Text(label="Result Text")
             file_output = gr.File()
             with gr.Row():
@@ -100,7 +127,7 @@ with gr.Blocks (theme="ParityError/Anime"  , css="#SUBMIT {background-color: #cd
         with gr.Column():
             image_output = gr.Image()
 
-    submit_btn.click(TextLineBox, input, outputs= [text_output, image_output, ] )
+    submit_btn.click(TextLineBox, inputs=[input,engine_input,], outputs= [text_output, image_output, ] )
     download_btn.click(Download, text_output, outputs= file_output )
     clear_btn.click(lambda: [None,None,None], inputs=None, outputs= [text_output, file_output, image_output])