Spaces:
Sleeping
Sleeping
| import os | |
| import gradio as gr | |
| import google.generativeai as genai | |
| # Configure Gemini API | |
| GEMINI_API_KEY = os.environ["GEMINI_API_KEY"] | |
| genai.configure(api_key=GEMINI_API_KEY) | |
| # Create the model with the same configuration as the sample | |
| generation_config = { | |
| "temperature": 1, | |
| "top_p": 0.95, | |
| "top_k": 40, | |
| "max_output_tokens": 8192, | |
| "response_mime_type": "text/plain", | |
| } | |
| model = genai.GenerativeModel( | |
| model_name="gemini-1.5-pro", | |
| generation_config=generation_config, | |
| system_instruction="You are an expert in detecting objects from xray image. Your job is to detect the objects from x-ray images.", | |
| ) | |
| def analyze_image(image, instruction): | |
| """Analyze the uploaded image using Gemini model""" | |
| try: | |
| if image is None: | |
| return "Please upload an image to analyze." | |
| # Use default instruction if none provided | |
| if not instruction: | |
| instruction = "Here is an xray image, describe all objects you can see in this image and their relative positions to each other." | |
| # Start a new chat session | |
| chat = model.start_chat() | |
| # Upload and analyze the image | |
| image_file = genai.upload_file(image) | |
| # Send the image with the provided instruction | |
| response = chat.send_message([ | |
| image_file, | |
| instruction | |
| ]) | |
| return response.text | |
| except Exception as e: | |
| return f"Error analyzing image: {str(e)}\nPlease make sure you've uploaded a valid image file." | |
| # Get example images from the examples directory | |
| examples_dir = os.path.join(os.path.dirname(__file__), "examples") | |
| example_images = [] | |
| if os.path.exists(examples_dir): | |
| example_images = [ | |
| [os.path.join(examples_dir, f), "Here is an xray image, do you see keys in this image, if yes then describe its nature in few words and location relative to the other objects."] | |
| for f in os.listdir(examples_dir) | |
| if f.lower().endswith(('.png', '.jpg', '.jpeg')) | |
| ] | |
| # Create Gradio interface | |
| iface = gr.Interface( | |
| fn=analyze_image, | |
| inputs=[ | |
| gr.Image(type="filepath", label="Upload X-ray Image"), | |
| gr.Textbox( | |
| label="Instruction", | |
| placeholder="Enter your instruction (e.g., 'do you see keys in this image? If yes, describe their location')", | |
| value="Here is an xray image, do you see keys in this image, if yes then describe its nature in few words and location relative to the other objects." | |
| ) | |
| ], | |
| outputs=gr.Textbox(label="Analysis Result", lines=10), | |
| title="X-ray Image Object Detection", | |
| description="Upload an X-ray image and provide instructions for analysis. The model will detect and describe objects based on your instructions.", | |
| examples=example_images, | |
| cache_examples=True | |
| ) | |
| # Launch the app | |
| if __name__ == "__main__": | |
| iface.launch() |