Update ocr_engine.py

ocr_engine.py

@@ -5,13 +5,11 @@ import re
 import logging
 from datetime import datetime
 import os
-from PIL import Image, ImageEnhance
-import pytesseract
 
-# Set up logging for detailed debugging
-logging.basicConfig(level=logging.DEBUG, format='%(asctime)s - %(levelname)s - %(message)s')
+# Set up logging for debugging
+logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
 
-# Initialize EasyOCR (enable GPU if available)
+# Initialize EasyOCR
 easyocr_reader = easyocr.Reader(['en'], gpu=False)
 
 # Directory for debug images
@@ -26,188 +24,326 @@ def save_debug_image(img, filename_suffix, prefix=""):
         cv2.imwrite(filename, img)
     else:  # Grayscale image
         cv2.imwrite(filename, img)
-    logging.debug(f"Saved debug image: {filename}")
+    logging.info(f"Saved debug image: {filename}")
 
 def estimate_brightness(img):
-    """Estimate image brightness to adjust processing"""
+    """Estimate image brightness to detect illuminated displays"""
     gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
-    brightness = np.mean(gray)
-    logging.debug(f"Estimated brightness: {brightness}")
-    return brightness
+    return np.mean(gray)
 
-def deblur_image(img):
-    """Apply iterative sharpening to reduce blur"""
-    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
-    # Multiple sharpening passes
-    for _ in range(2):
-        kernel = np.array([[-1, -1, -1], [-1, 9, -1], [-1, -1, -1]])
-        gray = cv2.filter2D(gray, -1, kernel)
-        gray = np.clip(gray, 0, 255).astype(np.uint8)
-    save_debug_image(gray, "00_deblurred")
-    return gray
-
-def preprocess_image(img):
-    """Enhance image for digit detection under adverse conditions"""
-    # PIL enhancement
-    pil_img = Image.fromarray(cv2.cvtColor(img, cv2.COLOR_BGR2RGB))
-    pil_img = ImageEnhance.Contrast(pil_img).enhance(3.0)  # Extreme contrast
-    pil_img = ImageEnhance.Brightness(pil_img).enhance(1.8)  # Strong brightness
-    img_enhanced = cv2.cvtColor(np.array(pil_img), cv2.COLOR_RGB2BGR)
-    save_debug_image(img_enhanced, "00_preprocessed_pil")
-
-    # Deblur
-    deblurred = deblur_image(img_enhanced)
-
-    # CLAHE for local contrast
-    clahe = cv2.createCLAHE(clipLimit=4.0, tileGridSize=(8, 8))
-    enhanced = clahe.apply(deblurred)
-    save_debug_image(enhanced, "00_clahe_enhanced")
-
-    # Noise reduction
-    filtered = cv2.bilateralFilter(enhanced, d=17, sigmaColor=200, sigmaSpace=200)
-    save_debug_image(filtered, "00_bilateral_filtered")
-
-    # Morphological cleaning
-    kernel = np.ones((5, 5), np.uint8)
-    filtered = cv2.morphologyEx(filtered, cv2.MORPH_OPEN, kernel, iterations=2)
-    save_debug_image(filtered, "00_morph_cleaned")
-    return filtered
-
-def normalize_image(img):
-    """Resize image to ensure digits are detectable"""
-    h, w = img.shape[:2]
-    target_height = 1080  # High resolution for small digits
-    aspect_ratio = w / h
-    target_width = int(target_height * aspect_ratio)
-    if target_width < 480:
-        target_width = 480
-        target_height = int(target_width / aspect_ratio)
-    resized = cv2.resize(img, (target_width, target_height), interpolation=cv2.INTER_CUBIC)
-    save_debug_image(resized, "00_normalized")
-    logging.debug(f"Normalized image to {target_width}x{target_height}")
-    return resized
-
-def tesseract_ocr(img):
-    """Fallback OCR using Tesseract"""
+def detect_roi(img):
+    """Detect and crop the region of interest (likely the digital display)"""
     try:
-        config = r'--oem 3 --psm 6 -c tessedit_char_whitelist=0123456789.-'
-        text = pytesseract.image_to_string(img, config=config).strip()
-        logging.info(f"Tesseract OCR raw text: {text}")
-        return text
+        save_debug_image(img, "01_original")
+        gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
+        save_debug_image(gray, "02_grayscale")
+        
+        # Use adaptive thresholding for better robustness
+        thresh = cv2.adaptiveThreshold(gray, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, 
+                                       cv2.THRESH_BINARY, 11, 2)
+        save_debug_image(thresh, "03_roi_adaptive_threshold")
+        
+        kernel = np.ones((7, 7), np.uint8)  # Smaller kernel
+        dilated = cv2.dilate(thresh, kernel, iterations=3)  # Fewer iterations
+        save_debug_image(dilated, "04_roi_dilated")
+        
+        contours, _ = cv2.findContours(dilated, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
+        
+        if contours:
+            img_area = img.shape[0] * img.shape[1]
+            valid_contours = []
+            for c in contours:
+                area = cv2.contourArea(c)
+                # Relaxed area and aspect ratio filters
+                if 500 < area < (img_area * 0.95):
+                    x, y, w, h = cv2.boundingRect(c)
+                    aspect_ratio = w / h
+                    if 1.5 <= aspect_ratio <= 6.0 and w > 80 and h > 40:
+                        valid_contours.append(c)
+
+            if valid_contours:
+                for contour in sorted(valid_contours, key=cv2.contourArea, reverse=True):
+                    x, y, w, h = cv2.boundingRect(contour)
+                    padding = 60  # Increased padding
+                    x, y = max(0, x - padding), max(0, y - padding)
+                    w, h = min(w + 2 * padding, img.shape[1] - x), min(h + 2 * padding, img.shape[0] - y)
+                    roi_img = img[y:y+h, x:x+w]
+                    save_debug_image(roi_img, "05_detected_roi")
+                    logging.info(f"Detected ROI with dimensions: ({x}, {y}, {w}, {h})")
+                    return roi_img, (x, y, w, h)
+        
+        logging.info("No suitable ROI found, returning original image.")
+        save_debug_image(img, "05_no_roi_original_fallback")
+        return img, None
     except Exception as e:
-        logging.error(f"Tesseract OCR failed: {str(e)}")
+        logging.error(f"ROI detection failed: {str(e)}")
+        save_debug_image(img, "05_roi_detection_error_fallback")
+        return img, None
+
+def detect_segments(digit_img):
+    """Detect seven-segment patterns in a digit image"""
+    h, w = digit_img.shape
+    if h < 15 or w < 10:
         return None
 
-def extract_weight_from_image(pil_img):
-    """Extract the actual weight shown in the image"""
-    try:
-        img = np.array(pil_img)
-        img = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
-        save_debug_image(img, "00_input_image")
+    segments = {
+        'top': (int(w*0.15), int(w*0.85), 0, int(h*0.2)),
+        'middle': (int(w*0.15), int(w*0.85), int(h*0.4), int(h*0.6)),
+        'bottom': (int(w*0.15), int(w*0.85), int(h*0.8), h),
+        'left_top': (0, int(w*0.25), int(h*0.05), int(h*0.5)),
+        'left_bottom': (0, int(w*0.25), int(h*0.5), int(h*0.95)),
+        'right_top': (int(w*0.75), w, int(h*0.05), int(h*0.5)),
+        'right_bottom': (int(w*0.75), w, int(h*0.5), int(h*0.95))
+    }
 
-        # Normalize image
-        img = normalize_image(img)
-        brightness = estimate_brightness(img)
-        conf_threshold = 0.1  # Very low threshold for blurry images
+    segment_presence = {}
+    for name, (x1, x2, y1, y2) in segments.items():
+        x1, y1 = max(0, x1), max(0, y1)
+        x2, y2 = min(w, x2), min(h, y2)
+        region = digit_img[y1:y2, x1:x2]
+        if region.size == 0:
+            segment_presence[name] = False
+            continue
+        pixel_count = np.sum(region == 255)
+        total_pixels = region.size
+        segment_presence[name] = pixel_count / total_pixels > 0.45  # Lowered threshold
 
-        # Preprocess entire image (bypass ROI detection)
-        processed_img = preprocess_image(img)
-        save_debug_image(processed_img, "01_processed_full")
+    digit_patterns = {
+        '0': ('top', 'bottom', 'left_top', 'left_bottom', 'right_top', 'right_bottom'),
+        '1': ('right_top', 'right_bottom'),
+        '2': ('top', 'middle', 'bottom', 'left_bottom', 'right_top'),
+        '3': ('top', 'middle', 'bottom', 'right_top', 'right_bottom'),
+        '4': ('middle', 'left_top', 'right_top', 'right_bottom'),
+        '5': ('top', 'middle', 'bottom', 'left_top', 'right_bottom'),
+        '6': ('top', 'middle', 'bottom', 'left_top', 'left_bottom', 'right_bottom'),
+        '7': ('top', 'right_top', 'right_bottom'),
+        '8': ('top', 'middle', 'bottom', 'left_top', 'left_bottom', 'right_top', 'right_bottom'),
+        '9': ('top', 'middle', 'bottom', 'left_top', 'right_top', 'right_bottom')
+    }
 
-        # Try multiple thresholding approaches
-        if brightness > 100:
-            thresh = cv2.adaptiveThreshold(processed_img, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, 
-                                           cv2.THRESH_BINARY_INV, 61, 11)
-            save_debug_image(thresh, "02_adaptive_threshold")
-        else:
-            _, thresh = cv2.threshold(processed_img, 10, 255, cv2.THRESH_BINARY_INV)
-            save_debug_image(thresh, "02_simple_threshold")
+    best_match = None
+    max_score = -1
+    for digit, pattern in digit_patterns.items():
+        matches = sum(1 for segment in pattern if segment_presence.get(segment, False))
+        non_matches_penalty = sum(1 for segment in segment_presence if segment not in pattern and segment_presence[segment])
+        current_score = matches - non_matches_penalty
+        if all(segment_presence.get(s, False) for s in pattern):
+            current_score += 0.5
+        if current_score > max_score:
+            max_score = current_score
+            best_match = digit
+        elif current_score == max_score and best_match is not None:
+            current_digit_non_matches = sum(1 for segment in segment_presence if segment not in pattern and segment_presence[segment])
+            best_digit_pattern = digit_patterns[best_match]
+            best_digit_non_matches = sum(1 for segment in segment_presence if segment not in best_digit_pattern and segment_presence[segment])
+            if current_digit_non_matches < best_digit_non_matches:
+                best_match = digit
+    
+    logging.debug(f"Segment presence: {segment_presence}, Detected digit: {best_match}")
+    return best_match
 
-        # Morphological operations
-        kernel = np.ones((7, 7), np.uint8)
-        thresh = cv2.morphologyEx(thresh, cv2.MORPH_CLOSE, kernel, iterations=3)
-        save_debug_image(thresh, "02_morph_cleaned")
+def custom_seven_segment_ocr(img, roi_bbox):
+    """Perform custom OCR for seven-segment displays"""
+    try:
+        gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
+        brightness = estimate_brightness(img)
+        if brightness > 150:
+            _, thresh = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
+        else:
+            _, thresh = cv2.threshold(gray, 80, 255, cv2.THRESH_BINARY)  # Lower threshold
+        save_debug_image(thresh, "06_roi_thresh_for_digits")
 
-        # EasyOCR attempt
         results = easyocr_reader.readtext(thresh, detail=1, paragraph=False, 
-                                         contrast_ths=0.05, adjust_contrast=1.5, 
-                                         text_threshold=0.05, mag_ratio=10.0, 
-                                         allowlist='0123456789.-', y_ths=0.8)
+                                          contrast_ths=0.2, adjust_contrast=0.8, 
+                                          text_threshold=0.7, mag_ratio=2.0, 
+                                          allowlist='0123456789.', y_ths=0.3)
         
         logging.info(f"EasyOCR results: {results}")
-        recognized_text = ""
-        if results:
-            # Sort by x-coordinate for left-to-right reading
-            sorted_results = sorted(results, key=lambda x: x[0][0][0])
-            for _, text, conf in sorted_results:
-                logging.info(f"EasyOCR detected: {text}, Confidence: {conf}")
-                if conf > conf_threshold and any(c in '0123456789.-' for c in text):
-                    recognized_text += text
-        else:
+        if not results:
             logging.info("EasyOCR found no digits.")
+            return None
 
-        if not recognized_text:
-            # Tesseract fallback
-            tesseract_result = tesseract_ocr(thresh)
-            if tesseract_result:
-                recognized_text = tesseract_result
-                logging.info(f"Using Tesseract result: {recognized_text}")
+        digits_info = []
+        for (bbox, text, conf) in results:
+            (x1, y1), (x2, y2), (x3, y3), (x4, y4) = bbox
+            h_bbox = max(y1, y2, y3, y4) - min(y1, y2, y3, y4)
+            if len(text) == 1 and (text.isdigit() or text == '.') and h_bbox > 8:
+                x_min, x_max = int(min(x1, x4)), int(max(x2, x3))
+                y_min, y_max = int(min(y1, y2)), int(max(y3, y4))
+                digits_info.append((x_min, x_max, y_min, y_max, text, conf))
 
-        logging.info(f"Raw recognized text: {recognized_text}")
-        if not recognized_text:
-            logging.info("No text detected by EasyOCR or Tesseract.")
-            return "Not detected", 0.0
-
-        # Minimal cleaning to preserve actual weight
-        text = recognized_text.lower().strip()
-        text = text.replace(",", ".").replace(";", ".").replace(":", ".").replace(" ", "")
-        text = text.replace("o", "0").replace("O", "0").replace("q", "0").replace("Q", "0")
-        text = text.replace("s", "5").replace("S", "5").replace("g", "9").replace("G", "6")
-        text = text.replace("l", "1").replace("I", "1").replace("|", "1")
-        text = text.replace("b", "8").replace("B", "8").replace("z", "2").replace("Z", "2")
-        text = text.replace("a", "4").replace("A", "4").replace("e", "3").replace("t", "7")
-        text = re.sub(r"(kgs|kg|k|lb|g|gr|pounds|lbs)\b", "", text)
-        text = re.sub(r"[^\d\.\-]", "", text)
+        digits_info.sort(key=lambda x: x[0])
+        recognized_text = ""
+        for idx, (x_min, x_max, y_min, y_max, easyocr_char, easyocr_conf) in enumerate(digits_info):
+            x_min, y_min = max(0, x_min), max(0, y_min)
+            x_max, y_max = min(thresh.shape[1], x_max), min(thresh.shape[0], y_max)
+            if x_max <= x_min or y_max <= y_min:
+                continue
+            digit_img_crop = thresh[y_min:y_max, x_min:x_max]
+            save_debug_image(digit_img_crop, f"07_digit_crop_{idx}_{easyocr_char}")
+            if easyocr_conf > 0.9 or easyocr_char == '.' or digit_img_crop.shape[0] < 15 or digit_img_crop.shape[1] < 10:
+                recognized_text += easyocr_char
+            else:
+                digit_from_segments = detect_segments(digit_img_crop)
+                if digit_from_segments:
+                    recognized_text += digit_from_segments
+                else:
+                    recognized_text += easyocr_char
         
+        logging.info(f"Before validation, recognized_text: {recognized_text}")
+        text = re.sub(r"[^\d\.]", "", recognized_text)
         if text.count('.') > 1:
-            parts = text.split('.')
-            text = parts[0] + '.' + ''.join(parts[1:])
-        text = text.strip('.')
-        
-        if text.startswith('.'):
-            text = "0" + text
-        if text.endswith('.'):
-            text = text.rstrip('.')
-
-        logging.info(f"Cleaned text: {text}")
-        if not text or text == '.' or text == '-':
-            logging.warning("Cleaned text is invalid.")
-            return "Not detected", 0.0
+            text = text.replace('.', '', text.count('.') - 1)
+        if text and re.fullmatch(r"^\d*\.?\d*$", text) and len(text) > 0:
+            if text.startswith('.'):
+                text = "0" + text
+            if text.endswith('.'):
+                text = text.rstrip('.')
+            if text == '.' or text == '':
+                return None
+            return text
+        logging.info(f"Custom OCR text '{recognized_text}' failed validation.")
+        return None
+    except Exception as e:
+        logging.error(f"Custom seven-segment OCR failed: {str(e)}")
+        return None
 
-        try:
-            weight = float(text)
-            confidence = 80.0 if recognized_text else 50.0
-            if weight < -1000 or weight > 2000:
-                logging.warning(f"Weight {weight} outside typical range, reducing confidence.")
-                confidence *= 0.5
-            if "." in text:
-                int_part, dec_part = text.split(".")
+def extract_weight_from_image(pil_img):
+    """Extract weight from a PIL image of a digital scale display"""
+    try:
+        img = np.array(pil_img)
+        img = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
+        save_debug_image(img, "00_input_image")  # Log input image
+
+        brightness = estimate_brightness(img)
+        conf_threshold = 0.6 if brightness > 150 else (0.5 if brightness > 80 else 0.4)
+
+        roi_img, roi_bbox = detect_roi(img)
+        custom_result = custom_seven_segment_ocr(roi_img, roi_bbox)
+        if custom_result:
+            if "." in custom_result:
+                int_part, dec_part = custom_result.split(".")
                 int_part = int_part.lstrip("0") or "0"
                 dec_part = dec_part.rstrip('0')
                 if not dec_part and int_part != "0":
-                    text = int_part
+                    custom_result = int_part
                 elif not dec_part and int_part == "0":
-                    text = "0"
+                    custom_result = "0"
                 else:
-                    text = f"{int_part}.{dec_part}"
+                    custom_result = f"{int_part}.{dec_part}"
             else:
-                text = text.lstrip('0') or "0"
-            logging.info(f"Final detected weight: {text}, Confidence: {confidence}%")
-            return text, confidence
-        except ValueError:
-            logging.warning(f"Could not convert '{text}' to float.")
+                custom_result = custom_result.lstrip('0') or "0"
+            try:
+                float(custom_result)
+                logging.info(f"Custom OCR result: {custom_result}, Confidence: 100.0%")
+                return custom_result, 100.0
+            except ValueError:
+                logging.warning(f"Custom OCR result '{custom_result}' is not a valid number, falling back.")
+                custom_result = None
+
+        logging.info("Custom OCR failed or invalid, falling back to general EasyOCR.")
+        processed_roi_img_gray = cv2.cvtColor(roi_img, cv2.COLOR_BGR2GRAY)
+        kernel_sharpening = np.array([[-1,-1,-1], [-1,9,-1], [-1,-1,-1]])
+        sharpened_roi = cv2.filter2D(processed_roi_img_gray, -1, kernel_sharpening)
+        save_debug_image(sharpened_roi, "08_fallback_sharpened")
+        processed_roi_img_final = cv2.adaptiveThreshold(sharpened_roi, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, 
+                                                       cv2.THRESH_BINARY, 21, 5)
+        save_debug_image(processed_roi_img_final, "09_fallback_adaptive_thresh")
+
+        results = easyocr_reader.readtext(processed_roi_img_final, detail=1, paragraph=False, 
+                                          contrast_ths=0.3, adjust_contrast=0.9, 
+                                          text_threshold=0.5, mag_ratio=2.0, 
+                                          allowlist='0123456789.', batch_size=4, y_ths=0.3)
+
+        best_weight = None
+        best_conf = 0.0
+        best_score = 0.0
+        for (bbox, text, conf) in results:
+            text = text.lower().strip()
+            text = text.replace(",", ".").replace(";", ".").replace(":", ".").replace(" ", "")
+            text = text.replace("o", "0").replace("O", "0").replace("q", "0").replace("Q", "0")
+            text = text.replace("s", "5").replace("S", "5")
+            text = text.replace("g", "9").replace("G", "6")
+            text = text.replace("l", "1").replace("I", "1").replace("|", "1")
+            text = text.replace("b", "8").replace("B", "8")
+            text = text.replace("z", "2").replace("Z", "2")
+            text = text.replace("a", "4").replace("A", "4")
+            text = text.replace("e", "3")
+            text = text.replace("t", "7")
+            text = text.replace("~", "").replace("`", "")
+            text = re.sub(r"(kgs|kg|k|lb|g|gr|pounds|lbs)\b", "", text)
+            text = re.sub(r"[^\d\.]", "", text)
+            if text.count('.') > 1:
+                parts = text.split('.')
+                text = parts[0] + '.' + ''.join(parts[1:])
+            text = text.strip('.')
+            if re.fullmatch(r"^\d*\.?\d{0,3}$", text) and len(text.replace('.', '')) > 0:
+                try:
+                    weight = float(text)
+                    range_score = 1.0
+                    if 0.1 <= weight <= 250:
+                        range_score = 1.5
+                    elif weight > 250 and weight <= 500:
+                        range_score = 1.2
+                    elif weight > 500 and weight <= 1000:
+                        range_score = 1.0
+                    else:
+                        range_score = 0.5
+                    digit_count = len(text.replace('.', ''))
+                    digit_score = 1.0
+                    if digit_count >= 2 and digit_count <= 5:
+                        digit_score = 1.3
+                    elif digit_count == 1:
+                        digit_score = 0.8
+                    score = conf * range_score * digit_score
+                    if roi_bbox:
+                        (x_roi, y_roi, w_roi, h_roi) = roi_bbox
+                        roi_area = w_roi * h_roi
+                        x_min, y_min = int(min(b[0] for b in bbox)), int(min(b[1] for b in bbox))
+                        x_max, y_max = int(max(b[0] for b in bbox)), int(max(b[1] for b in bbox))
+                        bbox_area = (x_max - x_min) * (y_max - y_min)
+                        if roi_area > 0 and bbox_area / roi_area < 0.03:
+                            score *= 0.5
+                        bbox_aspect_ratio = (x_max - x_min) / (y_max - y_min) if (y_max - y_min) > 0 else 0
+                        if bbox_aspect_ratio < 0.2:
+                            score *= 0.7
+                    if score > best_score and conf > conf_threshold:
+                        best_weight = text
+                        best_conf = conf
+                        best_score = score
+                        logging.info(f"Candidate EasyOCR weight: '{text}', Conf: {conf}, Score: {score}")
+                except ValueError:
+                    logging.warning(f"Could not convert '{text}' to float during EasyOCR fallback.")
+                    continue
+
+        if not best_weight:
+            logging.info("No valid weight detected after all attempts.")
             return "Not detected", 0.0
 
+        if "." in best_weight:
+            int_part, dec_part = best_weight.split(".")
+            int_part = int_part.lstrip("0") or "0"
+            dec_part = dec_part.rstrip('0')
+            if not dec_part and int_part != "0":
+                best_weight = int_part
+            elif not dec_part and int_part == "0":
+                best_weight = "0"
+            else:
+                best_weight = f"{int_part}.{dec_part}"
+        else:
+            best_weight = best_weight.lstrip('0') or "0"
+
+        try:
+            final_float_weight = float(best_weight)
+            if final_float_weight < 0.01 or final_float_weight > 1000:
+                logging.warning(f"Detected weight {final_float_weight} is outside typical range, reducing confidence.")
+                best_conf *= 0.5
+        except ValueError:
+            pass
+
+        logging.info(f"Final detected weight: {best_weight}, Confidence: {round(best_conf * 100, 2)}%")
+        return best_weight, round(best_conf * 100, 2)
+
     except Exception as e:
         logging.error(f"Weight extraction failed unexpectedly: {str(e)}")
         return "Not detected", 0.0