Update app.py

app.py

@@ -7,109 +7,138 @@ import gradio as gr
 from ultralytics import YOLO
 from paddleocr import PaddleOCR
 
-# 1) Load models
+# ─── 1) Load models ───────────────────────────────────────────────
 yolo = YOLO("models/best.pt")
-ocr  = PaddleOCR(
-    text_detection_model_dir=None,
-    text_recognition_model_dir="models/ocr_model",
+
+# Recognition-only PaddleOCR: no det_model_dir, so ocr.ocr() only runs the recognizer + cls
+ocr = PaddleOCR(
+    det_model_dir=None,               # disable PaddleOCR’s own detector entirely
+    rec_model_dir="models/ocr_model", # your trained CRNN weights
     use_textline_orientation=False,
-    lang="en"
+    lang="en",
+    use_angle_cls=True,               # flip/rotation correction
+    use_space_char=True               # allow spaces
 )
 
-# 2) Turkish plate formatter
+# ─── 2) Turkish plate formatter ────────────────────────────────────
 def format_turkish_plate(s: str) -> str:
     s = re.sub(r'[^A-Z0-9]', '', s.upper())
     m = re.match(r'^(\d{2})([A-Z]{1,3})(\d{2,4})$', s)
     return f"{m.group(1)} {m.group(2)} {m.group(3)}" if m else "Unknown"
 
-# 3) Single‐image inference
+# ─── 3) Flatten OCR result & pick min confidence ───────────────────
+def parse_ocr(recs):
+    """
+    recs from ocr.ocr(crop, cls=True) comes back as:
+      [  # list per text-line
+        [ <coords>, (<text>, <score>) ],
+        [ <coords>, (<text>, <score>) ],
+        ...
+      ]
+    We join all <text> pieces, and take the min score across them.
+    """
+    if not recs or not recs[0]:
+        return "", 0.0
+    lines = recs[0]
+    texts  = [line[1][0] for line in lines]
+    scores = [line[1][1] for line in lines]
+    return "".join(texts), float(min(scores))
+
+# ─── 4) Single-image inference ─────────────────────────────────────
 def run_image(img, conf=0.25):
+    # convert to BGR for YOLO
     bgr = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
-    res = yolo(bgr, conf=conf)[0]
+    res = yolo(bgr, conf=conf)[0]      # detect plates
     out = bgr.copy()
 
-    for box, yolo_score in zip(res.boxes.xyxy.cpu().numpy(),
-                               res.boxes.conf.cpu().numpy()):
-        x1,y1,x2,y2 = box.astype(int)
+    for box in res.boxes.xyxy.cpu().numpy().astype(int):
+        x1,y1,x2,y2 = box
         crop = out[y1:y2, x1:x2]
-        if crop.size==0: continue
-
-        plate_img = cv2.resize(crop, (128,32))
-        recs = ocr.ocr(plate_img, cls=False)
+        if crop.size == 0:
+            continue
 
-        if recs and recs[0]:
-            _, (raw, ocr_score) = recs[0]
-        else:
-            raw, ocr_score = "", 0.0
+        # resize for CRNN
+        plate_img = cv2.resize(crop, (128, 32))
+        # recognize only (no internal detection)
+        recs = ocr.ocr(plate_img, cls=True)
 
+        raw, score = parse_ocr(recs)
         plate = format_turkish_plate(raw)
-        label = f"{plate} ({ocr_score:.2f})"
+        label = f"{plate} ({score:.2f})"
 
-        cv2.rectangle(out, (x1,y1),(x2,y2), (0,255,0), 2)
-        cv2.putText(out, label, (x1,y1-5),
+        # draw box + label
+        cv2.rectangle(out, (x1,y1), (x2,y2), (0,255,0), 2)
+        cv2.putText(out, label, (x1, y1-5),
                     cv2.FONT_HERSHEY_SIMPLEX, 0.6, (0,255,0), 2)
 
     return cv2.cvtColor(out, cv2.COLOR_BGR2RGB), f"{len(res.boxes)} plate(s) detected"
 
-# 4) Video inference (same pattern)
+# ─── 5) Video inference ────────────────────────────────────────────
 def run_video(video_file, conf=0.25):
     cap = cv2.VideoCapture(video_file)
     fps = cap.get(cv2.CAP_PROP_FPS)
-    w,h = int(cap.get(3)), int(cap.get(4))
+    w   = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
+    h   = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
     outfp = tempfile.NamedTemporaryFile(suffix=".mp4", delete=False).name
     writer = cv2.VideoWriter(outfp, cv2.VideoWriter_fourcc(*"mp4v"), fps, (w,h))
-    records, idx = [], 0
 
+    records = []
+    frame_idx = 0
     while True:
         ret, frame = cap.read()
-        if not ret: break
-        idx += 1
-        t = idx / fps
+        if not ret:
+            break
+        frame_idx += 1
+        t = frame_idx / fps
 
         res = yolo(frame, conf=conf)[0]
         for (x1,y1,x2,y2) in res.boxes.xyxy.cpu().numpy().astype(int):
             crop = frame[y1:y2, x1:x2]
-            if crop.size==0: continue
-            plate_img = cv2.resize(crop, (128,32))
-            recs = ocr.ocr(plate_img, cls=False)
-
-            if recs and recs[0]:
-                _, (raw, ocr_score) = recs[0]
-            else:
-                raw, ocr_score = "", 0.0
+            if crop.size == 0:
+                continue
 
+            plate_img = cv2.resize(crop, (128,32))
+            recs = ocr.ocr(plate_img, cls=True)
+            raw, score = parse_ocr(recs)
             plate = format_turkish_plate(raw)
             if plate != "Unknown":
-                records.append({"time_s":round(t,2), "plate":plate, "conf":round(ocr_score,3)})
+                records.append({
+                    "time_s": round(t,2),
+                    "plate":  plate,
+                    "conf":   round(score,3)
+                })
 
             cv2.rectangle(frame, (x1,y1),(x2,y2), (0,255,0), 2)
-            cv2.putText(frame, plate, (x1,y1-5),
+            cv2.putText(frame, plate, (x1, y1-5),
                         cv2.FONT_HERSHEY_SIMPLEX, 0.6, (0,255,0), 2)
 
         writer.write(frame)
 
-    cap.release(); writer.release()
+    cap.release()
+    writer.release()
     with open("output.json","w") as f:
         json.dump(records, f, indent=2)
-    return outfp
 
-# 5) Gradio UI
+    return outfp, "Done"
+
+# ─── 6) Gradio UI ──────────────────────────────────────────────────
 with gr.Blocks() as demo:
     gr.Markdown("## 🚗 License Plate Detection + Recognition")
+
     with gr.Row():
         with gr.Column():
-            img_in = gr.Image(type="numpy")
-            vid_in = gr.File(label="Video (.mp4)")
-            conf   = gr.Slider(0,1,0.25,0.01)
+            img_in = gr.Image(type="numpy", label="Upload Image")
+            vid_in = gr.File(label="Upload Video (.mp4)")
+            conf   = gr.Slider(0,1,0.25,0.01, label="YOLO Confidence")
             b1     = gr.Button("Run Image")
             b2     = gr.Button("Run Video")
         with gr.Column():
-            img_out = gr.Image(type="numpy")
-            vid_out = gr.Video()
-            status  = gr.Textbox()
+            img_out = gr.Image(type="numpy", label="Annotated Image")
+            vid_out = gr.Video(label="Annotated Video")
+            status  = gr.Textbox(label="Status / JSON Path")
 
     b1.click(run_image, [img_in,conf], [img_out,status])
     b2.click(run_video, [vid_in,conf], [vid_out,status])
 
 if __name__=="__main__":
-    demo.launch()
+    demo.launch()