Upload 8 files

Procfile.txt

@@ -0,0 +1 @@
+web: gunicorn app:app

anpr.py

@@ -0,0 +1,229 @@
+import numpy as np
+import cv2
+import matplotlib.pyplot as plt
+import pytesseract as pt
+import os
+from transformers import TrOCRProcessor, VisionEncoderDecoderModel
+from PIL import Image
+import requests
+import torch
+import plotly.express as px
+
+# LOAD YOLO MODEL
+INPUT_WIDTH =  640
+INPUT_HEIGHT = 640
+#onnx_file_path = os.path.abspath('./static/models/best.onnx')
+onnx_file_path = os.path.abspath(r'./static/model/best.onnx')
+print(f"Attempting to load ONNX file from: {onnx_file_path}")
+processor = TrOCRProcessor.from_pretrained('microsoft/trocr-base-printed')
+model = VisionEncoderDecoderModel.from_pretrained('microsoft/trocr-base-printed') 
+
+
+
+if not os.path.exists(onnx_file_path):
+    print(f"Error: ONNX file not found at {onnx_file_path}")
+else:
+    try:
+        net = cv2.dnn.readNetFromONNX(onnx_file_path)
+        net.setPreferableBackend(cv2.dnn.DNN_BACKEND_OPENCV)
+        net.setPreferableTarget(cv2.dnn.DNN_TARGET_CPU)
+    except cv2.error as e:
+        print(f"Error loading ONNX file: {onnx_file_path}")
+        print(f"OpenCV error: {e}")
+        raise  # Re-raise the exception to halt the program
+
+
+
+
+
+def get_detections(img,net):
+    # CONVERT IMAGE TO YOLO FORMAT
+    image = img.copy()
+    row, col, d = image.shape
+
+    max_rc = max(row,col)
+    input_image = np.zeros((max_rc,max_rc,3),dtype=np.uint8)
+    input_image[0:row,0:col] = image
+
+    # GET PREDICTION FROM YOLO MODEL
+    blob = cv2.dnn.blobFromImage(input_image,1/255,(INPUT_WIDTH,INPUT_HEIGHT),swapRB=True,crop=False)
+    net.setInput(blob)
+    preds = net.forward()
+    detections = preds[0]
+    
+    return input_image, detections
+
+def non_maximum_supression(input_image,detections):
+    # FILTER DETECTIONS BASED ON CONFIDENCE AND PROBABILIY SCORE
+    # center x, center y, w , h, conf, proba
+    boxes = []
+    confidences = []
+
+    image_w, image_h = input_image.shape[:2]
+    x_factor = image_w/INPUT_WIDTH
+    y_factor = image_h/INPUT_HEIGHT
+
+    for i in range(len(detections)):
+        row = detections[i]
+        confidence = row[4] # confidence of detecting license plate
+        if confidence > 0.4:
+            class_score = row[5] # probability score of license plate
+            if class_score > 0.25:
+                cx, cy , w, h = row[0:4]
+
+                left = int((cx - 0.5*w)*x_factor)
+                top = int((cy-0.5*h)*y_factor)
+                width = int(w*x_factor)
+                height = int(h*y_factor)
+                box = np.array([left,top,width,height])
+
+                confidences.append(confidence)
+                boxes.append(box)
+
+    # clean
+    boxes_np = np.array(boxes).tolist()
+    confidences_np = np.array(confidences).tolist()
+    # NMS
+    index = np.array(cv2.dnn.NMSBoxes(boxes_np,confidences_np,0.25,0.45)).flatten()
+    
+    return boxes_np, confidences_np, index
+
+def extract_text_py(image,bbox):
+    x,y,w,h = bbox
+    
+    roi = image[y:y+h, x:x+w]
+    if 0 in roi.shape:
+        return ''
+    else:
+        roi_bgr = cv2.cvtColor(roi,cv2.COLOR_RGB2BGR)
+        gray = cv2.cvtColor(roi_bgr,cv2.COLOR_BGR2GRAY)
+        magic_color = apply_brightness_contrast(gray,brightness=40,contrast=70)
+        #text = pt.image_to_string(magic_color)
+        text = pt.image_to_string(magic_color,lang='eng',config='--psm 6')
+        text = text.strip()
+        
+        return text
+    
+# extrating text
+def extract_text(image,bbox):
+    x,y,w,h = bbox
+    roi = image[y:y+h, x:x+w]
+    #print("roi:",roi)
+
+    # Use OpenCV to read the image
+    img = roi.copy()
+    print(img.shape)
+    # Convert BGR to RGB
+    img_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
+
+    # Create the Plotly Express figure
+    #fig = px.imshow(img_rgb)
+
+    # Update layout and show the figure
+    #fig.update_layout(width=100, height=40, margin=dict(l=10, r=10, b=10, t=10))
+    #fig.update_xaxes(showticklabels=False).update_yaxes(showticklabels=False)
+    #fig.show()
+
+    image = img_rgb
+    if 0 in roi.shape:
+        return 'no number'
+
+    else:
+        pixel_values = processor(images=image, return_tensors="pt").pixel_values
+        generated_ids = model.generate(pixel_values)
+        text = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
+        text = filter_string(text)
+
+        return text
+    
+def filter_string(input_string):
+    filtered_chars = [char for char in input_string if char.isalnum() and (char.isupper() or char.isdigit())]
+    filtered_string = ''.join(filtered_chars)
+    return filtered_string
+  
+
+def drawings(image,boxes_np,confidences_np,index):
+    # drawings
+    text_list = []
+    for ind in index:
+        x,y,w,h =  boxes_np[ind]
+        bb_conf = confidences_np[ind]
+        conf_text = 'plate: {:.0f}%'.format(bb_conf*100)
+        license_text = extract_text(image,boxes_np[ind])
+
+
+        cv2.rectangle(image,(x,y),(x+w,y+h),(255,0,255),2)
+        cv2.rectangle(image,(x,y-30),(x+w,y),(255,0,255),-1)
+        cv2.rectangle(image,(x,y+h),(x+w,y+h+30),(0,0,0),-1)
+
+
+        cv2.putText(image,conf_text,(x,y-10),cv2.FONT_HERSHEY_SIMPLEX,0.7,(255,255,255),1)
+        cv2.putText(image,license_text,(x,y+h+27),cv2.FONT_HERSHEY_SIMPLEX,0.7,(0,255,0),1)
+        
+        text_list.append(license_text)
+
+    return image,  text_list
+
+
+# predictions
+def yolo_predictions(img,net):
+    ## step-1: detections
+    input_image, detections = get_detections(img,net)
+    ## step-2: NMS
+    boxes_np, confidences_np, index = non_maximum_supression(input_image, detections)
+    ## step-3: Drawings
+    result_img, text = drawings(img,boxes_np,confidences_np,index)
+    return result_img, text
+
+
+def object_detection(path,filename):
+    # read image
+    image = cv2.imread(path) # PIL object
+    image = np.array(image,dtype=np.uint8) # 8 bit array (0,255)
+    result_img, text_list = yolo_predictions(image,net)
+    cv2.imwrite('./static/predict/{}'.format(filename),result_img)
+    return text_list
+
+
+
+# def OCR(path,filename):
+#     img = np.array(load_img(path))
+#     cods = object_detection(path,filename)
+#     xmin ,xmax,ymin,ymax = cods[0]
+#     roi = img[ymin:ymax,xmin:xmax]
+#     roi_bgr = cv2.cvtColor(roi,cv2.COLOR_RGB2BGR)
+#     gray = cv2.cvtColor(roi_bgr,cv2.COLOR_BGR2GRAY)
+#     magic_color = apply_brightness_contrast(gray,brightness=40,contrast=70)
+#     cv2.imwrite('./static/roi/{}'.format(filename),roi_bgr)
+    
+    
+    
+#     print(text)
+#     save_text(filename,text)
+#     return text
+
+
+def apply_brightness_contrast(input_img, brightness = 0, contrast = 0):
+    
+        if brightness != 0:
+            if brightness > 0:
+                shadow = brightness
+                highlight = 255
+            else:
+                shadow = 0
+                highlight = 255 + brightness
+            alpha_b = (highlight - shadow)/255
+            gamma_b = shadow
+            
+            buf = cv2.addWeighted(input_img, alpha_b, input_img, 0, gamma_b)
+        else:
+            buf = input_img.copy()
+        
+        if contrast != 0:
+            f = 131*(contrast + 127)/(127*(131-contrast))
+            alpha_c = f
+            gamma_c = 127*(1-f)
+            
+            buf = cv2.addWeighted(buf, alpha_c, buf, 0, gamma_c)
+
+        return buf

app.py

@@ -0,0 +1,28 @@
+from flask import Flask, render_template, request
+import os 
+from anpr import object_detection
+# webserver gateway interface
+app = Flask(__name__)
+
+BASE_PATH = os.getcwd()
+UPLOAD_PATH = os.path.join(BASE_PATH,'static/upload/')
+
+
+@app.route('/',methods=['POST','GET'])
+def index():
+    if request.method == 'POST':
+        upload_file = request.files['image_name']
+        filename = upload_file.filename
+        path_save = os.path.join(UPLOAD_PATH,filename)
+        upload_file.save(path_save)
+        text_list = object_detection(path_save,filename)
+        
+        print(text_list)
+
+        return render_template('index.html',upload=True,upload_image=filename,text=text_list,no=len(text_list))
+
+    return render_template('index.html',upload=False)
+
+
+if __name__ =="__main__":
+    app.run(debug=True)

requirements.txt

@@ -0,0 +1,45 @@
+blinker==1.7.0
+brotlipy==0.7.0
+certifi==2021.5.30
+contourpy==1.2.0
+flask==3.0.0
+fonttools==4.47.2
+fsspec==2023.12.2
+geographiclib==1.52
+geopy==2.2.0
+gunicorn==19.9.0
+huggingface-hub==0.20.2
+importlib-metadata==7.0.1
+importlib-resources==6.1.1
+itsdangerous==2.1.2
+Jinja2==3.1.3
+kiwisolver==1.4.5
+MarkupSafe==2.1.3
+matplotlib==3.8.2
+mbstrdecoder==1.1.0
+mpmath==1.3.0
+networkx==3.2.1
+numpy==1.26.3
+opencv-python==4.9.0.80
+packaging==21.3
+pandas==2.2.0
+pillow==10.2.0
+plotly==5.18.0
+pycosat==0.6.3
+pyparsing==3.0.7
+pytesseract==0.3.10
+python-dateutil==2.8.2
+pytz==2022.1
+PyYAML==6.0.1
+regex==2023.12.25
+safetensors==0.4.1
+sympy==1.12
+tenacity==8.2.3
+tokenizers==0.15.0
+torch==2.1.2
+transformers==4.36.2
+typepy==1.3.0
+typing-extensions==4.9.0
+tzdata==2023.4
+werkzeug==3.0.1
+zipp==3.17.0

static/model/best.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:15c5cd92885be9bdab18f8a1d00d37ed037614a0c8a066d30bf62ce513455568
+size 28498982

static/model/best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:613dd4a3e709b45254dd5f0faa4c229f837ab4ba7f89cfb9b6cc2db4b6f7c890
+size 14327528

templates/index.html

@@ -0,0 +1,48 @@
+{% extends 'layout.html' %}
+
+{% block body %}
+    <div class="container">
+        <br><br>
+        <form action="#" method="POST" enctype="multipart/form-data">
+            <div class="input-group">
+                <input type="file" class="form-control" name="image_name" required>
+                <input type="submit" value="Upload" class="btn btn-outline-secondary">
+            </div>
+        </form>
+    </div>
+
+    {% if upload %}
+        <div class="container">
+            <br><br>
+            <table style="margin-bottom: 20px;"> <!-- Added margin-bottom for spacing -->
+                <tr style="border: solid black;">
+                    <th style="font-size: 20px;">Original Image</th>
+                    <th style="font-size: 20px;">Corresponding bounding box Image</th>
+                </tr>
+                <tr>
+                    <td>
+                        <img class="rounded float-left img-fluid" src="/static/upload/{{ upload_image }}" alt="">
+                    </td>
+                    <td>
+                        <img class="rounded float-right img-fluid" src="/static/predict/{{ upload_image }}" alt="">
+                    </td>
+                </tr>
+            </table>
+
+            <table style="border: solid black; width: 100%;">
+                <tr style="border: solid black; text-align: center;">
+                    <th style="font-size: 26px;">Number Plate of the Car</th>
+                </tr>
+                <tr style="border: solid black;">
+                    <td>
+                        <img class="img-fluid" src="/static/roi/{{ upload_image }}" alt="">
+                    </td>
+                    <td style="background-color: yellow;">
+                        <h1 class="display-2"> {{ text }}</h1>
+                    </td>
+                </tr>
+            </table>
+        </div>
+    {% endif %}
+
+{% endblock  %}

templates/layout.html

@@ -0,0 +1,62 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="UTF-8">
+    <meta http-equiv="X-UA-Compatible" content="IE=edge">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>ANPR Project</title>
+    <link href="https://cdn.jsdelivr.net/npm/[email protected]/dist/css/bootstrap.min.css" rel="stylesheet" integrity="sha384-T3c6CoIi6uLrA9TneNEoa7RxnatzjcDSCmG1MXxSR1GAsXEV/Dwwykc2MPK8M2HN" crossorigin="anonymous">
+    <script src="https://cdn.jsdelivr.net/npm/@popperjs/[email protected]/dist/umd/popper.min.js" integrity="sha384-I7E8VVD/ismYTF4hNIPjVp/Zjvgyol6VFvRkX/vR+Vc4jQkC+hVqc2pM8ODewa9r" crossorigin="anonymous"></script>
+    <script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/js/bootstrap.min.js" integrity="sha384-BBtl+eGJRgqQAUMxJ7pMwbEyER4l1g+O15P+16Ep7Q9Q+zqX6gSbd85u4mG4QzX+" crossorigin="anonymous"></script>
+    <!-- Add your CSS files or styles here -->
+</head>
+<body>
+    <div class="container">
+        <div class="row">
+            <div class="col-md-6">
+                <div class="col-md-6">
+                    <h2>Professor:</h2>
+                    <p>Ravi Kiran</p>
+                </div>
+                <h2>Members:</h2>
+                <ul>
+                    <li>Sreedhar Budamagunta</li>
+                    <li>Abhinav Govind Rao</li>
+                    <li>Praveen Srivastava</li>
+                </ul>
+            </div>
+            <div class="col-md-6">
+                <h2>Mentor:</h2>
+                <p>Sangeeth</p>
+            </div>
+        </div>
+    </div>
+
+    <nav class="navbar navbar-dark bg-dark">
+        <div class="container">
+            <a class="navbar-brand" href="/">
+                <h1 class="display-6">IIITH - Cohort 21 -  ANPR OCR Group 6 </h1>
+            </a>
+        </div>
+    </nav>
+    
+    {% block body %}
+    
+    {% endblock %}
+   
+
+    <footer>
+        <hr>
+        <h2>References</h2>
+        <a href="https://www.iiit.ac.in/" target="_blank">IIIT Hyderabad website</a>
+        <br>
+        <a href="https://pytorch.org/hub/ultralytics_yolov5/" target="_blank">YOLO5</a>
+        <br>
+        <a href="https://huggingface.co/docs/transformers/model_doc/trocr" target="_blank">Huggingface TrOCR</a>
+    </footer>
+    
+
+    <!-- Add your JavaScript files or scripts here -->
+
+</body>
+</html>