Component 01 it21269370 pothuwila p. y. r

Backend/.gitignore

@@ -0,0 +1 @@
+node_modules

Backend/arPoseLandmarks/downward_dog/downward_dog_landmarks.json

@@ -0,0 +1,42 @@
+[
+  {
+    "name": "left_shoulder",
+    "x": 0.43475762009620667,
+    "y": 0.5846027731895447
+  },
+  {
+    "name": "right_shoulder",
+    "x": 0.4466809630393982,
+    "y": 0.5982203483581543
+  },
+  {
+    "name": "left_elbow",
+    "x": 0.3332710564136505,
+    "y": 0.7483700513839722
+  },
+  {
+    "name": "right_elbow",
+    "x": 0.34916260838508606,
+    "y": 0.7354578375816345
+  },
+  {
+    "name": "left_hip",
+    "x": 0.6648672223091125,
+    "y": 0.2121756225824356
+  },
+  {
+    "name": "right_hip",
+    "x": 0.6585555076599121,
+    "y": 0.2347366362810135
+  },
+  {
+    "name": "left_knee",
+    "x": 0.8213318586349487,
+    "y": 0.516678512096405
+  },
+  {
+    "name": "right_knee",
+    "x": 0.7963972687721252,
+    "y": 0.5197720527648926
+  }
+]

Backend/arPoseLandmarks/extract_pose.py

@@ -0,0 +1,89 @@
+import cv2
+import mediapipe as mp
+import json
+import os
+
+print("[INFO] Initializing MediaPipe Pose...")
+mp_pose = mp.solutions.pose
+pose = mp_pose.Pose(
+    static_image_mode=True,
+    model_complexity=1,
+    min_detection_confidence=0.5,
+    min_tracking_confidence=0.5
+)
+mp_drawing = mp.solutions.drawing_utils
+
+# Update video path for Warrior II pose
+video_path = "../../CeylonCare/CeylonCare/assets/videos/warrior_ii.mp4"  # Relative path
+output_image_path = "warrior_ii/warrior_ii.jpg"
+output_json_path = "warrior_ii/warrior_ii_landmarks.json"
+
+# Create output directories if they don't exist
+os.makedirs(os.path.dirname(output_image_path), exist_ok=True)
+os.makedirs(os.path.dirname(output_json_path), exist_ok=True)
+
+if not os.path.exists(video_path):
+    print(f"[ERROR] Video file not found at: {video_path}")
+    print("[INFO] Ensure the video is in ../../CeylonCare/CeylonCare/assets/videos/")
+    exit()
+
+print("[INFO] Loading video...")
+cap = cv2.VideoCapture(video_path)
+if not cap.isOpened():
+    print("[ERROR] Could not open video.")
+    exit()
+
+fps = cap.get(cv2.CAP_PROP_FPS)
+total_frames = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
+print(f"[DEBUG] Video FPS: {fps}, Total Frames: {total_frames}")
+
+# Adjust target time based on when the Warrior II pose is held in the video
+target_time = 5  # Adjust this based on your video
+target_frame = int(fps * target_time) if int(fps * target_time) < total_frames else total_frames - 1
+print(f"[DEBUG] Targeting frame {target_frame} at {target_time} seconds")
+cap.set(cv2.CAP_PROP_POS_FRAMES, target_frame)
+
+success, frame = cap.read()
+if not success:
+    print("[ERROR] Failed to read frame.")
+    cap.release()
+    exit()
+
+print("[INFO] Processing frame for pose detection...")
+rgb_frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
+results = pose.process(rgb_frame)
+
+if not results.pose_landmarks:
+    print("[ERROR] No pose detected.")
+    cv2.imwrite(output_image_path, frame)
+    print(f"[DEBUG] Frame saved as {output_image_path}")
+    cap.release()
+    exit()
+
+mp_drawing.draw_landmarks(frame, results.pose_landmarks, mp_pose.POSE_CONNECTIONS)
+cv2.imwrite(output_image_path, frame)
+print(f"[INFO] Frame saved as {output_image_path}")
+
+# Extract landmarks
+landmarks = results.pose_landmarks.landmark
+correct_pose = [
+    {"name": "left_shoulder", "x": landmarks[11].x, "y": landmarks[11].y},
+    {"name": "right_shoulder", "x": landmarks[12].x, "y": landmarks[12].y},
+    {"name": "left_elbow", "x": landmarks[13].x, "y": landmarks[13].y},
+    {"name": "right_elbow", "x": landmarks[14].x, "y": landmarks[14].y},
+    {"name": "left_hip", "x": landmarks[23].x, "y": landmarks[23].y},
+    {"name": "right_hip", "x": landmarks[24].x, "y": landmarks[24].y},
+    {"name": "left_knee", "x": landmarks[25].x, "y": landmarks[25].y},
+    {"name": "right_knee", "x": landmarks[26].x, "y": landmarks[26].y}
+]
+print("[DEBUG] Extracted landmarks:")
+for landmark in correct_pose:
+    print(f"[DEBUG] {landmark['name']}: x={landmark['x']:.3f}, y={landmark['y']:.3f}")
+
+with open(output_json_path, 'w') as f:
+    json.dump(correct_pose, f, indent=2)
+print(f"[INFO] Landmarks saved to {output_json_path}")
+
+cap.release()
+pose.close()
+print("[INFO] Script completed!")

Backend/arPoseLandmarks/triangle_pose/triangle_pose_landmarks.json

@@ -0,0 +1,42 @@
+[
+  {
+    "name": "left_shoulder",
+    "x": 0.393909215927124,
+    "y": 0.5337827205657959
+  },
+  {
+    "name": "right_shoulder",
+    "x": 0.41303548216819763,
+    "y": 0.4024796485900879
+  },
+  {
+    "name": "left_elbow",
+    "x": 0.39781999588012695,
+    "y": 0.6615622043609619
+  },
+  {
+    "name": "right_elbow",
+    "x": 0.3966938853263855,
+    "y": 0.2836398184299469
+  },
+  {
+    "name": "left_hip",
+    "x": 0.50235915184021,
+    "y": 0.576718270778656
+  },
+  {
+    "name": "right_hip",
+    "x": 0.5291388630867004,
+    "y": 0.5406009554862976
+  },
+  {
+    "name": "left_knee",
+    "x": 0.4347962439060211,
+    "y": 0.7202959656715393
+  },
+  {
+    "name": "right_knee",
+    "x": 0.5943127870559692,
+    "y": 0.698548436164856
+  }
+]

Backend/arPoseLandmarks/warrior_ii/warrior_ii_landmarks.json

@@ -0,0 +1,42 @@
+[
+  {
+    "name": "left_shoulder",
+    "x": 0.5757824182510376,
+    "y": 0.30987370014190674
+  },
+  {
+    "name": "right_shoulder",
+    "x": 0.48430055379867554,
+    "y": 0.3139100670814514
+  },
+  {
+    "name": "left_elbow",
+    "x": 0.650972843170166,
+    "y": 0.32717323303222656
+  },
+  {
+    "name": "right_elbow",
+    "x": 0.40994828939437866,
+    "y": 0.3287721872329712
+  },
+  {
+    "name": "left_hip",
+    "x": 0.5621575117111206,
+    "y": 0.5378085374832153
+  },
+  {
+    "name": "right_hip",
+    "x": 0.5064576864242554,
+    "y": 0.5383180379867554
+  },
+  {
+    "name": "left_knee",
+    "x": 0.6146337985992432,
+    "y": 0.6881082057952881
+  },
+  {
+    "name": "right_knee",
+    "x": 0.44796282052993774,
+    "y": 0.6910121440887451
+  }
+]

Backend/ar_model/pose_service.py

@@ -0,0 +1,69 @@
+from flask import Flask, request, jsonify
+import cv2
+import mediapipe as mp
+import base64
+import numpy as np
+import logging
+import traceback
+
+# Set up logging for debugging and error tracking
+logging.basicConfig(level=logging.DEBUG)
+logger = logging.getLogger(__name__)
+
+app = Flask(__name__)
+mp_pose = mp.solutions.pose
+
+# Initialize MediaPipe Pose model with higher confidence thresholds
+pose = mp_pose.Pose(min_detection_confidence=0.7, min_tracking_confidence=0.7)
+
+@app.route('/process_frame', methods=['POST'])
+def process_frame():
+    try:
+        # Log incoming request
+        logger.info('[DEBUG] Received request to process frame')
+
+        # Get JSON data from request
+        data = request.json
+        if not data or 'frame' not in data:
+            logger.error('[ERROR] Invalid request: Missing frame data')
+            return jsonify({'error': 'Missing frame data'}), 400
+
+        # Extract and log frame data
+        frame_base64 = data['frame']
+        logger.debug('[DEBUG] Frame base64 length: %d', len(frame_base64))
+
+        # Decode base64 string to image
+        frame = base64.b64decode(frame_base64)
+        image = cv2.imdecode(np.frombuffer(frame, np.uint8), cv2.IMREAD_COLOR)
+        if image is None:
+            logger.error('[ERROR] Failed to decode image')
+            return jsonify({'error': 'Failed to decode image'}), 400
+
+        # Log image details
+        logger.debug('[DEBUG] Image decoded, shape: %s', image.shape)
+
+        # Convert image to RGB for MediaPipe processing
+        image_rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
+        results = pose.process(image_rgb)
+
+        # Process pose detection results
+        if results.pose_landmarks:
+            landmarks = [
+                {'name': f'landmark_{i}', 'x': lm.x, 'y': lm.y, 'z': lm.z}
+                for i, lm in enumerate(results.pose_landmarks.landmark)
+            ]
+            logger.info('[DEBUG] Pose landmarks detected: %d', len(landmarks))
+            return jsonify({'landmarks': landmarks})
+        else:
+            logger.warning('[WARN] No pose landmarks detected')
+            return jsonify({'landmarks': []})
+
+    except Exception as e:
+        # Log detailed error information
+        logger.error('[ERROR] Error processing frame: %s', str(e))
+        logger.error('[ERROR] Traceback: %s', traceback.format_exc())
+        return jsonify({'error': str(e)}), 500
+
+if __name__ == '__main__':
+    logger.info('[INFO] Starting pose detection service on http://0.0.0.0:5002')
+    app.run(host='0.0.0.0', port=5002, debug=True)

Backend/controllers/arController.js

@@ -14,7 +14,7 @@ const getARRecommendations = async (req, res) => {
   console.log(`[DEBUG] Received request to fetch therapy for user: ${userId}`);
 
   try {
-    const response = await axios.get(`http://localhost:5000/healthData/${userId}`);
+    const response = await axios.get(`http://192.168.60.22:5000/healthData/${userId}`);
     console.log(`[DEBUG] Fetching health data for user: ${userId}, Response status: ${response.status}`);
 
     if (response.status !== 200) {
@@ -24,7 +24,7 @@ const getARRecommendations = async (req, res) => {
     const userHealthData = response.data;
     console.log("[DEBUG] Fetched user health data:", JSON.stringify(userHealthData, null, 2));
 
-    const flaskResponse = await axios.post("http://localhost:5001/predict", userHealthData);
+    const flaskResponse = await axios.post("http://192.168.60.22:5001/predict", userHealthData);
 
     console.log("[DEBUG] Flask API Full Response:", JSON.stringify(flaskResponse.data, null, 2));
 

Backend/controllers/chatController.js

@@ -33,7 +33,7 @@ const getChatRecommendation = async (req, res) => {
     console.log(`[DEBUG] User health condition: ${userHealthData.healthCondition || 'general'}`);
 
     // Send chat message and health condition to Flask app
-    const flaskResponse = await axios.post("http://localhost:5001/chat", {
+    const flaskResponse = await axios.post("http://192.168.60.22:5001/chat", {
       message: userInput,
       userId: userId,
       healthCondition: userHealthData.healthCondition || "general",