Update app.py

app.py

@@ -163,144 +163,244 @@ def generate_report(
     io_times: List[float]
 ) -> str:
     log_entries.append("Generating report...")
-    report_path = os.path.join(OUTPUT_DIR, f"drone_analysis_report_{datetime.now().strftime('%Y%m%d_%H%M%S')}.md")
+    report_path = os.path.join(OUTPUT_DIR, f"drone_analysis_report_{datetime.now().strftime('%Y%m%d_%H%M%S')}.tex")
     timestamp = datetime.now().strftime('%Y%m%d_%H%M%S')
-    report_content = [
-        "# NHAI Drone Survey Analysis Report",
-        "",
-        "## Project Details",
-        "- Project Name: NH-44 Delhi-Hyderabad Section (Package XYZ)",
-        "- Highway Section: Km 100 to Km 150",
-        "- State: Telangana",
-        "- Region: South",
-        f"- Survey Date: {datetime.now().strftime('%Y-%m-%d')}",
-        "- Drone Service Provider: ABC Drone Services Pvt. Ltd.",
-        "- Technology Service Provider: XYZ AI Analytics Ltd.",
-        f"- Work Order Reference: Data Lake WO-{datetime.now().strftime('%Y-%m-%d')}-XYZ",
-        "- Report Prepared By: Nagasurendra, Data Analyst",
-        f"- Report Date: {datetime.now().strftime('%Y-%m-%d')}",
-        "",
-        "## 1. Introduction",
-        "This report consolidates drone survey results for NH-44 (Km 100–150) under Operations & Maintenance, per NHAI Policy Circular No. 18.98/2024, detecting potholes and cracks using YOLOv8 for Monthly Progress Report integration.",
-        "",
-        "## 2. Drone Survey Metadata",
-        "- Drone Speed: 5 m/s",
-        "- Drone Height: 60 m",
-        "- Camera Sensor: RGB, 12 MP",
-        "- Recording Type: JPEG, 90° nadir",
-        "- Image Overlap: 85%",
-        "- Flight Pattern: Single lap, ROW centered",
-        "- Geotagging: Enabled",
-        "- Satellite Lock: 12 satellites",
-        "- Terrain Follow Mode: Enabled",
-        "",
-        "## 3. Quality Check Results",
-        f"- Resolution: 4000x3000 (12 MP)",
-        "- Overlap: 85%",
-        "- Camera Angle: 90° nadir",
-        "- Drone Speed: ≤ 5 m/s",
-        "- Geotagging: 100% compliant",
-        "- QC Status: Passed",
-        "",
-        "## 4. AI/ML Analytics",
-        f"- Total Frames Processed: {frame_count}",
-        f"- Detection Frames: {detection_frame_count} ({detection_frame_count/frame_count*100:.2f}%)",
-        f"- Total Detections: {metrics['total_detections']}",
-        "  - Breakdown:"
-    ]
 
+    # Prepare items for the breakdown
+    items = []
     for item in metrics.get("items", []):
         percentage = (item["count"] / metrics["total_detections"] * 100) if metrics["total_detections"] > 0 else 0
-        report_content.append(f"    - {item['type']}: {item['count']} ({percentage:.2f}%)")
-    report_content.extend([
-        f"- Processing Time: {total_time:.2f} seconds",
-        f"- Average Frame Time: {sum(frame_times)/len(frame_times):.2f} ms" if frame_times else "- Average Frame Time: N/A",
-        f"- Average Resize Time: {sum(resize_times)/len(resize_times):.2f} ms" if resize_times else "- Average Resize Time: N/A",
-        f"- Average Inference Time: {sum(inference_times)/len(inference_times):.2f} ms" if inference_times else "- Average Inference Time: N/A",
-        f"- Average I/O Time: {sum(io_times)/len(io_times):.2f} ms" if io_times else "- Average I/O Time: N/A",
-        f"- Timestamp: {metrics.get('timestamp', 'N/A')}",
-        "- Summary: Potholes and cracks detected in high-traffic segments.",
-        "",
-        "## 5. Output File Structure",
-        "- ZIP file contains:",
-        "  - `drone_analysis_report_<timestamp>.md`: This report",
-        "  - `outputs/processed_output.mp4`: Processed video with annotations",
-        "  - `outputs/chart_<timestamp>.png`: Detection trend chart",
-        "  - `outputs/map_<timestamp>.png`: Issue locations map",
-        "  - `captured_frames/detected_<frame>.jpg`: Geotagged images for detected issues",
-        "  - `flight_logs/flight_log_<frame>.csv`: Flight logs matching image frames",
-        "- Note: Images and logs share frame numbers (e.g., `detected_000001.jpg` corresponds to `flight_log_000001.csv`).",
-        "",
-        "## 6. Geotagged Images",
-        f"- Total Images: {len(detected_issues)}",
-        f"- Storage: Data Lake `/project_xyz/images/{datetime.now().strftime('%Y-%m-%d')}`",
-        "",
-        "| Frame | Issue Type | GPS (Lat, Lon) | Timestamp | Confidence | Image Path |",
-        "|-------|------------|----------------|-----------|------------|------------|"
-    ])
+        items.append(f"\\item {item['type']}: {item['count']} ({percentage:.2f}\\%)")
 
+    # Prepare image rows with embedded images
+    image_rows = []
     for detection in all_detections[:100]:
-        report_content.append(
-            f"| {detection['frame']:06d} | {detection['label']} | ({detection['gps'][0]:.6f}, {detection['gps'][1]:.6f}) | {detection['timestamp']} | {detection['conf']:.2f} | captured_frames/{os.path.basename(detection['path'])} |"
+        image_basename = os.path.basename(detection['path'])
+        image_rows.append(
+            f"{detection['frame']:06d} & {detection['label']} & "
+            f"({detection['gps'][0]:.6f}, {detection['gps'][1]:.6f}) & "
+            f"{detection['timestamp']} & {detection['conf']:.2f} & "
+            f"\\includegraphics[width=4cm, height=3cm, keepaspectratio]{{{image_basename}}} \\\\"
         )
 
-    report_content.extend([
-        "",
-        "## 7. Flight Logs",
-        f"- Total Logs: {len(detected_issues)}",
-        f"- Storage: Data Lake `/project_xyz/flight_logs/{datetime.now().strftime('%Y-%m-%d')}`",
-        "",
-        "| Frame | Timestamp | Latitude | Longitude | Speed (m/s) | Satellites | Altitude (m) | Log Path |",
-        "|-------|-----------|----------|-----------|-------------|------------|--------------|----------|"
-    ])
-
+    # Prepare flight log rows
+    log_rows = []
     for detection in all_detections[:100]:
         log_path = f"flight_logs/flight_log_{detection['frame']:06d}.csv"
-        report_content.append(
-            f"| {detection['frame']:06d} | {detection['timestamp']} | {detection['gps'][0]:.6f} | {detection['gps'][1]:.6f} | 5.0 | 12 | 60 | {log_path} |"
+        log_rows.append(
+            f"{detection['frame']:06d} & {detection['timestamp']} & "
+            f"{detection['gps'][0]:.6f} & {detection['gps'][1]:.6f} & "
+            f"5.0 & 12 & 60 & \\texttt{{{log_path}}} \\\\"
         )
 
-    report_content.extend([
-        "",
-        "## 8. Processed Video",
-        f"- Path: outputs/processed_output.mp4",
-        f"- Frames: {output_frames}",
-        f"- FPS: {output_fps:.2f}",
-        f"- Duration: {output_duration:.2f} seconds",
-        "",
-        "## 9. Visualizations",
-        f"- Detection Trend Chart: outputs/chart_{timestamp}.png",
-        f"- Issue Locations Map: outputs/map_{timestamp}.png",
-        "",
-        "## 10. Processing Timestamps",
-        f"- Total Processing Time: {total_time:.2f} seconds",
-        "- Log Entries (Last 10):"
-    ])
-
+    # Prepare log entries
+    log_entries_str = []
     for entry in log_entries[-10:]:
-        report_content.append(f"  - {entry}")
-
-    report_content.extend([
-        "",
-        "## 11. Stakeholder Validation",
-        "- AE/IE Comments: [Pending]",
-        "- PD/RO Comments: [Pending]",
-        "",
-        "## 12. Recommendations",
-        "- Repair potholes in high-traffic segments.",
-        "- Seal cracks to prevent degradation.",
-        "- Schedule follow-up survey.",
-        "",
-        "## 13. Data Lake References",
-        f"- Images: `/project_xyz/images/{datetime.now().strftime('%Y-%m-%d')}`",
-        f"- Flight Logs: `/project_xyz/flight_logs/{datetime.now().strftime('%Y-%m-%d')}`",
-        f"- Video: `/project_xyz/videos/processed_output_{datetime.now().strftime('%Y%m%d')}.mp4`",
-        f"- DAMS Dashboard: `/project_xyz/dams/{datetime.now().strftime('%Y-%m-%d')}`"
-    ])
+        log_entries_str.append(f"\\item {entry}")
+
+    # Calculate average times
+    avg_frame_time = sum(frame_times) / len(frame_times) if frame_times else 0
+    avg_resize_time = sum(resize_times) / len(resize_times) if resize_times else 0
+    avg_inference_time = sum(inference_times) / len(inference_times) if inference_times else 0
+    avg_io_time = sum(io_times) / len(io_times) if io_times else 0
+
+    # LaTeX template
+    latex_template = r"""
+\documentclass[a4paper,12pt]{article}
+\usepackage[utf8]{inputenc}
+\usepackage[T1]{fontenc}
+\usepackage{geometry}
+\geometry{margin=1in}
+\usepackage{graphicx}
+\usepackage{booktabs}
+\usepackage{longtable}
+\usepackage{hyperref}
+\usepackage{xcolor}
+\usepackage{parskip}
+\setlength{\parskip}{0.5em}
+\usepackage{noto}
+\graphicspath{{./captured_frames/}}
+
+\begin{document}
+
+% Title and project details
+\section*{NHAI Drone Survey Analysis Report}
+
+\subsection*{Project Details}
+\begin{itemize}
+    \item \textbf{Project Name:} NH-44 Delhi-Hyderabad Section (Package XYZ)
+    \item \textbf{Highway Section:} Km 100 to Km 150
+    \item \textbf{State:} Telangana
+    \item \textbf{Region:} South
+    \item \textbf{Survey Date:} \today
+    \item \textbf{Drone Service Provider:} ABC Drone Services Pvt. Ltd.
+    \item \textbf{Technology Service Provider:} XYZ AI Analytics Ltd.
+    \item \textbf{Work Order Reference:} Data Lake WO-\today-XYZ
+    \item \textbf{Report Prepared By:} Nagasurendra, Data Analyst
+    \item \textbf{Report Date:} \today
+\end{itemize}
+
+\section{Introduction}
+This report consolidates drone survey results for NH-44 (Km 100--150) under Operations \& Maintenance, per NHAI Policy Circular No. 18.98/2024, detecting potholes and cracks using YOLOv8 for Monthly Progress Report integration.
+
+\section{Drone Survey Metadata}
+\begin{itemize}
+    \item \textbf{Drone Speed:} 5 m/s
+    \item \textbf{Drone Height:} 60 m
+    \item \textbf{Camera Sensor:} RGB, 12 MP
+    \item \textbf{Recording Type:} JPEG, 90° nadir
+    \item \textbf{Image Overlap:} 85\%
+    \item \textbf{Flight Pattern:} Single lap, ROW centered
+    \item \textbf{Geotagging:} Enabled
+    \item \textbf{Satellite Lock:} 12 satellites
+    \item \textbf{Terrain Follow Mode:} Enabled
+\end{itemize}
+
+\section{Quality Check Results}
+\begin{itemize}
+    \item \textbf{Resolution:} 4000x3000 (12 MP)
+    \item \textbf{Overlap:} 85\%
+    \item \textbf{Camera Angle:} 90° nadir
+    \item \textbf{Drone Speed:} $\leq$ 5 m/s
+    \item \textbf{Geotagging:} 100\% compliant
+    \item \textbf{QC Status:} Passed
+\end{itemize}
+
+\section{AI/ML Analytics}
+\begin{itemize}
+    \item \textbf{Total Frames Processed:} {frame_count}
+    \item \textbf{Detection Frames:} {detection_frame_count} ({detection_percentage:.2f}\%)
+    \item \textbf{Total Detections:} {total_detections}
+    \item \textbf{Breakdown:}
+    \begin{itemize}
+        {items}
+    \end{itemize}
+    \item \textbf{Processing Time:} {total_time:.2f} seconds
+    \item \textbf{Average Frame Time:} {avg_frame_time:.2f} ms
+    \item \textbf{Average Resize Time:} {avg_resize_time:.2f} ms
+    \item \textbf{Average Inference Time:} {avg_inference_time:.2f} ms
+    \item \textbf{Average I/O Time:} {avg_io_time:.2f} ms
+    \item \textbf{Timestamp:} {timestamp_str}
+    \item \textbf{Summary:} Potholes and cracks detected in high-traffic segments.
+\end{itemize}
+
+\section{Output File Structure}
+\begin{itemize}
+    \item ZIP file contains:
+    \begin{itemize}
+        \item \texttt{drone_analysis_report_{timestamp}.pdf}: This report
+        \item \texttt{outputs/processed_output.mp4}: Processed video with annotations
+        \item \texttt{outputs/chart_{timestamp}.png}: Detection trend chart
+        \item \texttt{outputs/map_{timestamp}.png}: Issue locations map
+        \item \texttt{captured_frames/detected_<frame>.jpg}: Geotagged images for detected issues
+        \item \texttt{flight_logs/flight_log_<frame>.csv}: Flight logs matching image frames
+    \end{itemize}
+    \item \textbf{Note:} Images and logs share frame numbers (e.g., \texttt{detected_000001.jpg} corresponds to \texttt{flight_log_000001.csv}).
+\end{itemize}
+
+\section{Geotagged Images}
+\begin{itemize}
+    \item \textbf{Total Images:} {total_images}
+    \item \textbf{Storage:} Data Lake \texttt{{/project_xyz/images/{date_str}}}
+\end{itemize}
+
+\begin{longtable}{c l c c c p{4cm}}
+    \toprule
+    \textbf{Frame} & \textbf{Issue Type} & \textbf{GPS (Lat, Lon)} & \textbf{Timestamp} & \textbf{Confidence} & \textbf{Image} \\
+    \midrule
+    \endhead
+    {image_rows}
+    \bottomrule
+\end{longtable}
+
+\section{Flight Logs}
+\begin{itemize}
+    \item \textbf{Total Logs:} {total_logs}
+    \item \textbf{Storage:} Data Lake \texttt{{/project_xyz/flight_logs/{date_str}}}
+\end{itemize}
+
+\begin{longtable}{c c c c c c c l}
+    \toprule
+    \textbf{Frame} & \textbf{Timestamp} & \textbf{Latitude} & \textbf{Longitude} & \textbf{Speed (m/s)} & \textbf{Satellites} & \textbf{Altitude (m)} & \textbf{Log Path} \\
+    \midrule
+    \endhead
+    {log_rows}
+    \bottomrule
+\end{longtable}
+
+\section{Processed Video}
+\begin{itemize}
+    \item \textbf{Path:} \texttt{outputs/processed_output.mp4}
+    \item \textbf{Frames:} {output_frames}
+    \item \textbf{FPS:} {output_fps:.2f}
+    \item \textbf{Duration:} {output_duration:.2f} seconds
+\end{itemize}
+
+\section{Visualizations}
+\begin{itemize}
+    \item \textbf{Detection Trend Chart:} \texttt{outputs/chart_{timestamp}.png}
+    \item \textbf{Issue Locations Map:} \texttt{outputs/map_{timestamp}.png}
+\end{itemize}
+
+\section{Processing Timestamps}
+\begin{itemize}
+    \item \textbf{Total Processing Time:} {total_time:.2f} seconds
+    \item \textbf{Log Entries (Last 10):}
+    \begin{itemize}
+        {log_entries}
+    \end{itemize}
+\end{itemize}
+
+\section{Stakeholder Validation}
+\begin{itemize}
+    \item \textbf{AE/IE Comments:} [Pending]
+    \item \textbf{PD/RO Comments:} [Pending]
+\end{itemize}
+
+\section{Recommendations}
+\begin{itemize}
+    \item Repair potholes in high-traffic segments.
+    \item Seal cracks to prevent degradation.
+    \item Schedule follow-up survey.
+\end{itemize}
+
+\section{Data Lake References}
+\begin{itemize}
+    \item \textbf{Images:} \texttt{{/project_xyz/images/{date_str}}}
+    \item \textbf{Flight Logs:} \texttt{{/project_xyz/flight_logs/{date_str}}}
+    \item \textbf{Video:} \texttt{{/project_xyz/videos/processed_output_{date_str_no_dash}.mp4}}
+    \item \textbf{DAMS Dashboard:} \texttt{{/project_xyz/dams/{date_str}}}
+\end{itemize}
+
+\end{document}
+"""
+
+    # Format the LaTeX template
+    report_content = latex_template.format(
+        frame_count=frame_count,
+        detection_frame_count=detection_frame_count,
+        detection_percentage=(detection_frame_count / frame_count * 100) if frame_count > 0 else 0,
+        total_detections=metrics['total_detections'],
+        items="\n        ".join(items),
+        total_time=total_time,
+        avg_frame_time=avg_frame_time,
+        avg_resize_time=avg_resize_time,
+        avg_inference_time=avg_inference_time,
+        avg_io_time=avg_io_time,
+        timestamp_str=metrics.get('timestamp', 'N/A'),
+        timestamp=timestamp,
+        total_images=len(detected_issues),
+        total_logs=len(detected_issues),
+        date_str=datetime.now().strftime('%Y-%m-%d'),
+        date_str_no_dash=datetime.now().strftime('%Y%m%d'),
+        image_rows="\n    ".join(image_rows),
+        log_rows="\n    ".join(log_rows),
+        log_entries="\n        ".join(log_entries_str)
+    )
 
     try:
         with open(report_path, 'w') as f:
-            f.write("\n".join(report_content))
+            f.write(report_content)
         log_entries.append(f"Report saved: {report_path}")
         return report_path
     except Exception as e:
@@ -415,6 +515,7 @@ def process_video(video, resize_width=4000, resize_height=3000, frame_skip=5):
 
         io_times.append((time.time() - io_start) * 1000)
 
+ torrents
         out.write(annotated_frame)
         output_frame_count += 1
         last_annotated_frame = annotated_frame
@@ -504,7 +605,7 @@ with gr.Blocks(theme=gr.themes.Soft(primary_hue="orange")) as iface:
         video_output = gr.Video(label="Processed Video")
         issue_gallery = gr.Gallery(label="Detected Issues", columns=4, height="auto", object_fit="contain")
     with gr.Row():
-        chart_output = gr.Image(label="Detection Trend")
+        chart_output = gr.Image(label-gate="Detection Trend")
         map_output = gr.Image(label="Issue Locations Map")
     with gr.Row():
         logs_output = gr.Textbox(label="Logs", lines=5, interactive=False)