Merge branch 'feature/IT20224820/pest_detection' of...

Merge branch 'feature/IT20224820/pest_detection' of http://gitlab.sliit.lk/2023-113/2023-113 into feature/IT20224820/pest_detection

Pest Identier and Protection Suggestioner-IT20224820/Flask server/class_mapping

+class_mapping = {
+    0: 'Ash weevil',
+    1: 'Blister Beetle',
+    2: 'Fruit fly',
+    3: 'Fruit sucking moth',
+    4: 'Helicoverpa',
+    5: 'Hellula Undalis',
+    6: 'Leaf Webber',
+    7: 'Leucinodes',
+    8: 'Mealy Bug',
+    9: 'Plutella',
+    10: 'Root Grubs',
+    11: 'Schizaphis graminum',
+    12: 'Uroleucon compositae',
+    13: 'pieris',
+    14: 'whitefly',
+}
\ No newline at end of file

Pest Identier and Protection Suggestioner-IT20224820/Flask server/main.py

+import io
+import time
+from flask import Flask, request, jsonify
+from PIL import Image
+import torch
+from flask_cors import CORS
+import requests
+import base64
+import numpy as np
+
+app = Flask(__name__)
+
+CORS(app)
+CORS(app, resources={r"/*": {"origins": "*"}})
+
+# Load your trained YOLOv5 model
+model = torch.hub.load('ultralytics/yolov5', 'custom', path='./best.pt')
+model.eval()
+
+# Get the class names
+class_names = model.names if hasattr(model, 'names') else model.module.names
+
+@app.route('/detect', methods=['POST'])
+def detect_objects():
+    image_file = request.files['image']
+    image = Image.open(image_file)
+
+    # Run inference using your trained model
+    start_time = time.time()
+    results = model(image)
+    end_time = time.time()
+
+    # Process the inference results
+    detections = results.pandas().xyxy[0]  # Get the detections as a pandas DataFrame
+
+    # Extract the relevant information
+    image_shape = image.size  # Get the shape of the input image
+    num_objects = len(detections)  # Get the number of detected objects
+    inference_time = end_time - start_time  # Calculate the inference time
+
+    # Extract class names and confidence scores
+    class_ids_detected = detections['class'].tolist()
+    confidence_scores = detections['confidence'].tolist()
+
+    # Draw bounding boxes on the image
+    annotated_image = results.show()
+
+    # Prepare the output response
+    response = {
+        'image_path': image_file.filename,
+        'image_shape': image_shape,
+        'num_objects': num_objects,
+        'inference_time': inference_time,
+        'detections': [{'class_name': class_names[class_id], 'confidence': confidence}
+                       for class_id, confidence in zip(class_ids_detected, confidence_scores)],
+        'annotated_image': annotated_image
+    }
+
+    # Return the response as JSON
+    return jsonify(response)
+
+@app.route('/link', methods=['POST'])
+def detect_objects_link():
+    # Get the image URL from the request
+    image_url = request.json['image_url']
+
+    # Download the image from the URL
+    response = requests.get(image_url)
+    image = Image.open(io.BytesIO(response.content))
+
+    # Convert image to numpy array
+    image_np = np.array(image)
+
+    # Run inference using your trained model
+    results = model(image_np)
+
+    # Process the inference results
+    annotated_image = results.render()[0]
+
+    # Convert the annotated image to PIL Image
+    annotated_image_pil = Image.fromarray(annotated_image)
+
+    # Save the annotated image to a file
+    annotated_image_path = './save/annotated_image.jpg'
+    annotated_image_pil.save(annotated_image_path)
+
+    # Convert the annotated image to base64
+    with open(annotated_image_path, 'rb') as image_file:
+        base64_image = base64.b64encode(image_file.read()).decode('utf-8')
+
+    # Prepare the output response
+    response = {
+        'image_url': image_url,
+        'num_objects': len(results.pandas().xyxy[0]),
+        'detections': [{'class_name': class_names[class_id], 'confidence': confidence}
+                       for class_id, confidence in zip(results.pandas().xyxy[0]['class'].tolist(),
+                                                       results.pandas().xyxy[0]['confidence'].tolist())],
+        'annotated_image': base64_image
+    }
+
+    # Return the response as JSON
+    return jsonify(response)
+
+if __name__ == '__main__':
+    app.run()
\ No newline at end of file

Pest Identier and Protection Suggestioner-IT20224820/Flask server/requirements.txt

+
+flask==2.3.2
+torch==2.0.1
+flask_cors==3.0.10
+requests==2.30.0
+numpy==1.24.3
\ No newline at end of file