qudi MCP Integration

#!/usr/bin/env python3 """ Simple transmission spectrum extractor using PIL Extracts approximate data points from transmission.png for RKHS analysis """ import numpy as np from PIL import Image import matplotlib.pyplot as plt import json def extract_spectrum_from_image(image_path, wavelength_range=(400, 700), transmission_range=(0, 100)): """ Extract spectrum data from image using simple pixel analysis. This is a basic approach - for production use, consider using axiomatic-plots MCP server. """ # Load image img = Image.open(image_path) img_array = np.array(img) print(f"Image dimensions: {img_array.shape}") # Convert to grayscale if color image if len(img_array.shape) == 3: # Convert to grayscale using standard weights gray = np.dot(img_array[...,:3], [0.2989, 0.5870, 0.1140]) else: gray = img_array height, width = gray.shape print(f"Processing grayscale image: {width} x {height}") # Find the darkest path (spectrum line) in each column x_pixels = [] y_pixels = [] # Look for the darkest points in each column (assuming dark spectrum line on light background) for x in range(width): column = gray[:, x] # Find the darkest point in this column min_idx = np.argmin(column) # Only include if it's significantly darker than average if column[min_idx] < np.mean(column) - np.std(column): x_pixels.append(x) y_pixels.append(min_idx) if len(x_pixels) < 10: # Try alternative approach - look for edges print("Trying edge-based approach...") edges = np.abs(np.diff(gray, axis=0)) x_pixels = [] y_pixels = [] for x in range(width): column_edges = edges[:, x] if x < edges.shape[1] else [] if len(column_edges) > 0: # Find strongest edge max_edge_idx = np.argmax(column_edges) if column_edges[max_edge_idx] > np.mean(column_edges) + np.std(column_edges): x_pixels.append(x) y_pixels.append(max_edge_idx) if len(x_pixels) < 10: raise ValueError(f"Could not extract sufficient data points from image. Only found {len(x_pixels)} points.") # Convert pixel coordinates to physical units x_pixels = np.array(x_pixels) y_pixels = np.array(y_pixels) # Map x pixels to wavelength wavelengths = wavelength_range[0] + (x_pixels / width) * (wavelength_range[1] - wavelength_range[0]) # Map y pixels to transmission (flip y-axis since image coordinates are top-to-bottom) transmission = transmission_range[1] - (y_pixels / height) * (transmission_range[1] - transmission_range[0]) # Sort by wavelength sort_idx = np.argsort(wavelengths) wavelengths = wavelengths[sort_idx] transmission = transmission[sort_idx] # Remove duplicates by averaging nearby points unique_wavelengths = [] unique_transmission = [] tolerance = (wavelength_range[1] - wavelength_range[0]) / width * 2 # 2-pixel tolerance i = 0 while i < len(wavelengths): # Find all points within tolerance close_indices = np.abs(wavelengths - wavelengths[i]) < tolerance # Average them avg_wavelength = np.mean(wavelengths[close_indices]) avg_transmission = np.mean(transmission[close_indices]) unique_wavelengths.append(avg_wavelength) unique_transmission.append(avg_transmission) # Skip processed points i = np.max(np.where(close_indices)[0]) + 1 return np.array(unique_wavelengths), np.array(unique_transmission) def plot_extracted_data(wavelengths, transmission, save_path=None): """Plot the extracted spectrum data""" plt.figure(figsize=(12, 6)) plt.plot(wavelengths, transmission, 'o-', markersize=3, linewidth=1) plt.xlabel('Wavelength (nm)') plt.ylabel('Transmission (%)') plt.title('Extracted Transmission Spectrum Data') plt.grid(True, alpha=0.3) if save_path: plt.savefig(save_path, dpi=300, bbox_inches='tight') print(f"✓ Extracted data plot saved to: {save_path}") return plt.gcf() def main(): """Main execution""" image_path = '/Users/englund/Projects/2025-ai-experiments-MCPs/plot_extraction/transmission.png' try: print("Extracting spectrum data from transmission.png...") wavelengths, transmission = extract_spectrum_from_image( image_path, wavelength_range=(400, 700), # Adjust based on actual spectrum range transmission_range=(0, 100) # Adjust based on actual transmission range ) print(f"✓ Extracted {len(wavelengths)} data points") print(f"Wavelength range: {wavelengths.min():.1f} - {wavelengths.max():.1f} nm") print(f"Transmission range: {transmission.min():.1f} - {transmission.max():.1f}%") # Look for double peak structure around 500-550nm peak_region = (wavelengths >= 500) & (wavelengths <= 550) if np.any(peak_region): peak_wavelengths = wavelengths[peak_region] peak_transmissions = transmission[peak_region] print(f"Double peak region (500-550nm): {len(peak_wavelengths)} points") print(f"Peak transmission range: {peak_transmissions.min():.1f} - {peak_transmissions.max():.1f}%") # Plot extracted data plot_extracted_data(wavelengths, transmission, '/Users/englund/Projects/2025-ai-experiments-MCPs/plot_extraction/extracted_spectrum.png') # Save extracted data data = { 'wavelengths': wavelengths.tolist(), 'transmission': transmission.tolist(), 'extraction_method': 'simple_pil_based', 'image_source': image_path, 'data_points': len(wavelengths), 'wavelength_range': [float(wavelengths.min()), float(wavelengths.max())], 'transmission_range': [float(transmission.min()), float(transmission.max())] } with open('/Users/englund/Projects/2025-ai-experiments-MCPs/plot_extraction/extracted_data.json', 'w') as f: json.dump(data, f, indent=2) print(f"✓ Data saved to extracted_data.json") return wavelengths, transmission except Exception as e: print(f"❌ Error extracting data: {e}") return None, None if __name__ == "__main__": main()