Upload 16 files

.gitattributes

@@ -0,0 +1,5 @@
+data/alert_response.mp4 filter=lfs diff=lfs merge=lfs -text
+data/drone_day.mp4 filter=lfs diff=lfs merge=lfs -text
+data/night_intrusion.mp4 filter=lfs diff=lfs merge=lfs -text
+data/shadow_dust_issue.mp4 filter=lfs diff=lfs merge=lfs -text
+data/thermal_hotspot.mp4 filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,14 @@
+---
+title: Solar Surveillance Poc
+emoji: 📚
+colorFrom: red
+colorTo: yellow
+sdk: gradio
+sdk_version: 5.26.0
+app_file: app.py
+pinned: false
+license: apache-2.0
+short_description: Solar Panels Suveillance using Tech & AI
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,140 @@
+import gradio as gr
+import cv2
+import time
+import os
+import random
+import matplotlib.pyplot as plt
+import numpy as np
+from datetime import datetime
+from services.video_service import get_next_video_frame, reset_video_index
+from services.thermal_service import detect_thermal_anomalies
+from services.overlay_service import overlay_boxes
+from services.metrics_service import update_metrics
+
+# Globals
+paused = False
+frame_rate = 1
+frame_count = 0
+log_entries = []
+anomaly_counts = []
+last_frame = None
+last_metrics = {}
+last_timestamp = ""
+last_detected_images = []
+
+# Constants
+TEMP_IMAGE_PATH = "temp.jpg"
+CAPTURED_FRAMES_DIR = "captured_frames"
+os.makedirs(CAPTURED_FRAMES_DIR, exist_ok=True)
+
+# Core monitor function
+def monitor_feed():
+    global paused, frame_count, last_frame, last_metrics, last_timestamp
+
+    if paused and last_frame is not None:
+        frame = last_frame.copy()
+        metrics = last_metrics.copy()
+    else:
+        frame = get_next_video_frame()
+        detected_boxes = detect_thermal_anomalies(frame)
+        frame = overlay_boxes(frame, detected_boxes)
+        cv2.imwrite(TEMP_IMAGE_PATH, frame, [int(cv2.IMWRITE_JPEG_QUALITY), 95])
+        metrics = update_metrics(detected_boxes)
+
+        frame_count += 1
+        last_timestamp = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
+
+        if detected_boxes:
+            captured_frame_path = os.path.join(CAPTURED_FRAMES_DIR, f"frame_{frame_count}.jpg")
+            cv2.imwrite(captured_frame_path, frame)
+            last_detected_images.append(captured_frame_path)
+            if len(last_detected_images) > 5:
+                last_detected_images.pop(0)
+
+        last_frame = frame.copy()
+        last_metrics = metrics.copy()
+
+    frame = cv2.resize(last_frame, (640, 480))
+    cv2.putText(frame, f"Frame: {frame_count}", (10, 25), cv2.FONT_HERSHEY_SIMPLEX, 0.6, (0, 255, 0), 2)
+    cv2.putText(frame, f"{last_timestamp}", (10, 50), cv2.FONT_HERSHEY_SIMPLEX, 0.6, (0, 255, 0), 2)
+
+    anomaly_detected = len(last_metrics.get('anomalies', []))
+    log_entries.append(f"{last_timestamp} - Frame {frame_count} - Anomalies: {anomaly_detected}")
+    anomaly_counts.append(anomaly_detected)
+
+    if len(log_entries) > 100:
+        log_entries.pop(0)
+   # if len(anomaly_counts) > 100:
+    #    anomaly_counts.pop(0)
+
+    metrics_str = "\n".join([f"{k}: {v}" for k, v in last_metrics.items()])
+
+    return frame[:, :, ::-1], metrics_str, "\n".join(log_entries[-10:]), generate_chart(), last_detected_images
+
+# Chart generator
+def generate_chart():
+    fig, ax = plt.subplots(figsize=(4, 2))
+    ax.plot(anomaly_counts[-50:], marker='o')
+    ax.set_title("Anomalies Over Time")
+    ax.set_xlabel("Frame")
+    ax.set_ylabel("Count")
+    fig.tight_layout()
+    chart_path = "chart_temp.png"
+    fig.savefig(chart_path)
+    plt.close(fig)
+    return chart_path
+
+# Gradio UI
+with gr.Blocks(theme=gr.themes.Soft()) as app:
+    gr.Markdown("# 🌐 Thermal Anomaly Monitoring Dashboard", elem_id="main-title")
+
+    status_text = gr.Markdown("**Status:** 🟢 Running", elem_id="status-banner")
+
+    with gr.Row():
+        with gr.Column(scale=3):
+            video_output = gr.Image(label="Live Video Feed", elem_id="video-feed", width=640, height=480)
+        with gr.Column(scale=1):
+            metrics_output = gr.Textbox(label="Live Metrics", lines=5)
+
+    with gr.Row():
+        with gr.Column():
+            logs_output = gr.Textbox(label="Live Logs", lines=10)
+        with gr.Column():
+            chart_output = gr.Image(label="Detection Trends")
+
+    with gr.Row():
+        captured_images = gr.Gallery(label="Last 5 Captured Events", columns=1, height="auto")
+
+    with gr.Row():
+        pause_btn = gr.Button("⏸️ Pause")
+        resume_btn = gr.Button("▶️ Resume")
+        frame_slider = gr.Slider(0.0005, 0.5, value=1, label="Frame Interval (seconds)")
+
+    def toggle_pause():
+        global paused
+        paused = True
+        return "**Status:** ⏸️ Paused"
+
+    def toggle_resume():
+        global paused
+        paused = False
+        return "**Status:** 🟢 Running"
+
+    def set_frame_rate(val):
+        global frame_rate
+        frame_rate = val
+
+    pause_btn.click(toggle_pause, outputs=status_text)
+    resume_btn.click(toggle_resume, outputs=status_text)
+    frame_slider.change(set_frame_rate, inputs=[frame_slider])
+
+    def streaming_loop():
+        while True:
+            frame, metrics, logs, chart, captured = monitor_feed()
+            yield frame, metrics, logs, chart, captured
+            time.sleep(frame_rate)
+
+    app.load(streaming_loop, outputs=[video_output, metrics_output, logs_output, chart_output, captured_images])
+
+if __name__ == "__main__":
+    app.launch(share=True)

data/alert_response.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90be2a22be37eec0ff56fc17b15c3472c91581b08276bd9c986143ce65199ecb
+size 2874687

data/drone_day.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02277520b32f06d332647c11687e8642efb95b7808a37c5e8d62fb289e128b0e
+size 97974586

data/night_intrusion.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6bc3952af22c9688512f650e668996e0e328a04c3f407b73860f1181da610f8
+size 2974359

data/shadow_dust_issue.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:acd094f281fa7fb35ff8425afb7f820a30be44cc9450709d854a0271ea35038b
+size 231805841

data/thermal_hotspot.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e2e2fcc90f9a894ed226192936b92170f9f77b635c2cf257c61f16f51fe0b18
+size 3349808

gitattributes

@@ -0,0 +1,45 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+services/alert_response.mp4 filter=lfs diff=lfs merge=lfs -text
+services/drone_day.mp4 filter=lfs diff=lfs merge=lfs -text
+services/night_intrusion.mp4 filter=lfs diff=lfs merge=lfs -text
+services/shadow_dust_issue.mp4 filter=lfs diff=lfs merge=lfs -text
+services/thermal_hotspot.mp4 filter=lfs diff=lfs merge=lfs -text
+data/alert_response.mp4 filter=lfs diff=lfs merge=lfs -text
+data/drone_day.mp4 filter=lfs diff=lfs merge=lfs -text
+data/night_intrusion.mp4 filter=lfs diff=lfs merge=lfs -text
+data/shadow_dust_issue.mp4 filter=lfs diff=lfs merge=lfs -text
+data/thermal_hotspot.mp4 filter=lfs diff=lfs merge=lfs -text

requirements.txt

@@ -0,0 +1,8 @@
+gradio==4.10.0
+transformers==4.39.3
+torch>=2.1.0
+opencv-python-headless
+requests
+ultralytics==8.0.176
+timm>=0.9.2
+

services/detection_service.py

@@ -0,0 +1,37 @@
+'''
+from transformers import pipeline
+import cv2
+
+object_detector = pipeline("object-detection", model="facebook/detr-resnet-50")
+
+def detect_objects(image_path):
+    image = cv2.imread(image_path)
+    results = object_detector(image)
+    return [r for r in results if r['score'] > 0.7]
+
+'''
+
+# services/detection_service.py
+
+from transformers import pipeline
+from PIL import Image
+
+# ✅ Load Hugging Face DETR pipeline properly
+object_detector = pipeline("object-detection", model="facebook/detr-resnet-50")
+
+def detect_objects(image_path):
+    """
+    Detect objects using Hugging Face DETR pipeline.
+    - Accepts a file path to a local image.
+    - Converts image to PIL format.
+    - Feeds into Hugging Face detector.
+    - Returns list of high-confidence detections.
+    """
+    # ✅ Correct way: Open image properly
+    image = Image.open(image_path).convert("RGB")
+
+    # ✅ Run inference
+    results = object_detector(image)
+
+    # ✅ Return only high-confidence detections
+    return [r for r in results if r['score'] > 0.7]

services/metrics_service.py

@@ -0,0 +1,6 @@
+def update_metrics(detected_boxes):
+    anomalies = detected_boxes if detected_boxes else []
+    return {
+        "anomalies": anomalies,
+        "total_detected": len(anomalies),
+    }

services/overlay_service.py

@@ -0,0 +1,7 @@
+import cv2
+
+def overlay_boxes(frame, boxes):
+    for box in boxes:
+        x_min, y_min, x_max, y_max = map(int, box)
+        cv2.rectangle(frame, (x_min, y_min), (x_max, y_max), (0, 0, 255), 2)
+    return frame

services/salesforce_dispatcher.py

@@ -0,0 +1,24 @@
+# services/salesforce_dispatcher.py
+
+import requests
+import json
+
+SALESFORCE_WEBHOOK_URL = "https://your-salesforce-instance/services/web-to-case"
+
+def send_to_salesforce(payload):
+    """
+    FAKE Salesforce integration — DISABLED in HuggingFace Space demo.
+    Replace SALESFORCE_WEBHOOK_URL later with real endpoint during production setup.
+    """
+    print("🚀 [Salesforce Dispatch Simulated] Would have sent:", payload)
+    # Commenting actual POST request to avoid Space crash
+    #
+    # alert_type = "Intrusion" if any(d["label"] == "person" for d in payload["detections"]) else "Anomaly"
+    # summary = {
+    #     "Alert_Type__c": alert_type,
+    #     "ThermalFlag__c": payload["thermal"],
+    #     "ShadowFlag__c": payload["shadow_issue"],
+    #     "Confidence_Score__c": max([d["score"] for d in payload["detections"]], default=0)
+    # }
+    # headers = {"Content-Type": "application/json"}
+    # requests.post(SALESFORCE_WEBHOOK_URL, json=summary, headers=headers)

services/shadow_detection.py

@@ -0,0 +1,5 @@
+import random
+
+def detect_shadow_coverage(image_path):
+    shadow_percent = random.randint(25, 40)
+    return shadow_percent > 30

services/thermal_service.py

@@ -0,0 +1,23 @@
+from transformers import DetrImageProcessor, DetrForObjectDetection
+import torch
+from PIL import Image
+import cv2
+
+# Load model
+processor = DetrImageProcessor.from_pretrained("facebook/detr-resnet-50")
+model = DetrForObjectDetection.from_pretrained("facebook/detr-resnet-50")
+
+def detect_thermal_anomalies(frame):
+    image = Image.fromarray(cv2.cvtColor(frame, cv2.COLOR_BGR2RGB))
+    inputs = processor(images=image, return_tensors="pt")
+    outputs = model(**inputs)
+
+    target_sizes = torch.tensor([image.size[::-1]])
+    results = processor.post_process_object_detection(outputs, threshold=0.9, target_sizes=target_sizes)[0]
+
+    boxes = []
+    for score, label, box in zip(results["scores"], results["labels"], results["boxes"]):
+        if score >= 0.9:
+            boxes.append(box.tolist())
+
+    return boxes

services/video_service.py

@@ -0,0 +1,33 @@
+import cv2
+import os
+
+# Global state
+VIDEO_DIR = "data"
+video_files = [os.path.join(VIDEO_DIR, file) for file in sorted(os.listdir(VIDEO_DIR)) if file.endswith((".mp4", ".avi"))]
+video_index = 0
+cap = None
+
+def get_next_video_frame():
+    global cap, video_index
+
+    if not video_files:
+        raise RuntimeError("No video files found in the 'data' directory.")
+
+    if cap is None or not cap.isOpened():
+        cap = cv2.VideoCapture(video_files[video_index])
+
+    ret, frame = cap.read()
+
+    if not ret:
+        cap.release()
+        video_index = (video_index + 1) % len(video_files)
+        cap = cv2.VideoCapture(video_files[video_index])
+        ret, frame = cap.read()
+        if not ret:
+            raise RuntimeError(f"Cannot read video {video_files[video_index]}")
+
+    return frame
+
+def reset_video_index():
+    global video_index
+    video_index = 0