add files

app.py

@@ -0,0 +1,72 @@
+import streamlit as st
+from akaocr import TextEngine, BoxEngine
+import cv2
+import numpy as np
+from PIL import Image
+import time
+
+# Initialize the OCR engines
+box_engine = BoxEngine()
+text_engine = TextEngine()
+
+def transform_image(image, box):
+    # Get perspective transform image
+    assert len(box) == 4, "Shape of points must be 4x2"
+    img_crop_width = int(
+        max(
+            np.linalg.norm(box[0] - box[1]),
+            np.linalg.norm(box[2] - box[3])))
+    img_crop_height = int(
+        max(
+            np.linalg.norm(box[0] - box[3]),
+            np.linalg.norm(box[1] - box[2])))
+    pts_std = np.float32([[0, 0], 
+                        [img_crop_width, 0],
+                        [img_crop_width, img_crop_height],
+                        [0, img_crop_height]])
+    box = np.array(box, dtype="float32")
+    M = cv2.getPerspectiveTransform(box, pts_std)
+    dst_img = cv2.warpPerspective(
+        image,
+        M, (img_crop_width, img_crop_height),
+        borderMode=cv2.BORDER_REPLICATE,
+        flags=cv2.INTER_CUBIC)
+    
+    img_height, img_width = dst_img.shape[0:2]
+    if img_height/img_width >= 1.25:
+            dst_img = np.rot90(dst_img, k=3)
+
+    return dst_img
+
+def main():
+    st.title("OCR Application with akaOCR")
+
+    uploaded_file = st.file_uploader("Choose an image...", type=["jpg", "jpeg", "png"])
+
+    if uploaded_file is not None:
+        # Convert the uploaded file to an OpenCV image
+        file_bytes = np.asarray(bytearray(uploaded_file.read()), dtype=np.uint8)
+        org_image = cv2.imdecode(file_bytes, 1)
+        st.image(org_image, channels="BGR", caption='Uploaded Image')
+
+        images = []
+        start = time.perf_counter()
+        boxes = box_engine(org_image)
+        processing_time = time.perf_counter() - start
+        st.write(f"Box detection took {processing_time:.2f} seconds.")
+
+        for box in boxes:
+            org_image = cv2.polylines(org_image, [box.astype(np.int32)], isClosed=True, color=(0, 255, 0), thickness=2)
+            image = transform_image(org_image, box)
+            images.append(image)
+
+        texts = text_engine(images)
+
+        # Convert back to PIL Image for displaying
+        output_image = Image.fromarray(cv2.cvtColor(org_image, cv2.COLOR_BGR2RGB))
+        st.image(output_image, caption='Detected Text Boxes', use_column_width=True)
+        st.write("Extracted Texts:")
+        st.write(texts)
+
+if __name__ == '__main__':
+    main()

requirements.txt

@@ -0,0 +1,4 @@
+streamlit
+akaocr
+opencv-python-headless
+numpy