dockerize the app

Dockerfile

@@ -0,0 +1,31 @@
+# HF Spaces compatible base image
+FROM python:3.9-slim
+
+# Prevent Python from writing .pyc files
+ENV PYTHONDONTWRITEBYTECODE=1
+ENV PYTHONUNBUFFERED=1
+
+# Set working directory
+WORKDIR /app
+
+# System dependencies (safe defaults)
+RUN apt-get update && apt-get install -y \
+    git \
+    curl \
+    && rm -rf /var/lib/apt/lists/*
+
+# Copy requirements first (better Docker caching)
+COPY requirements.txt .
+
+# Upgrade pip and install dependencies
+RUN pip install --upgrade pip \
+    && pip install --no-cache-dir -r requirements.txt
+
+# Copy the rest of the application
+COPY . .
+
+# Hugging Face Spaces uses port 7860
+EXPOSE 7860
+
+# Start the app
+CMD ["python", "app.py"]

README.md

@@ -3,7 +3,7 @@ title: ImageRecognition
 emoji: 💬
 colorFrom: yellow
 colorTo: purple
-sdk: gradio
+sdk: docker
 sdk_version: 5.0.1
 app_file: app.py
 pinned: false

app.py

@@ -1,384 +1,3 @@
-# import os
-# os.environ["CUDA_VISIBLE_DEVICES"] = "-1"
-
-
-# import os
-# import tempfile
-# import numpy as np
-# import cv2
-# import gradio as gr
-# from tensorflow.keras.applications import ResNet50
-# from tensorflow.keras.applications.resnet50 import preprocess_input
-# from tensorflow.keras.preprocessing import image
-# from skimage.metrics import structural_similarity as ssim
-# from PIL import Image
-# from io import BytesIO
-
-
-# # Disable GPU for TensorFlow
-# os.environ["TF_ENABLE_ONEDNN_OPTS"] = "0"
-# os.environ["CUDA_VISIBLE_DEVICES"] = "-1"
-
-# class ImageCharacterClassifier:
-#     def __init__(self, similarity_threshold=0.5):
-#         self.model = ResNet50(weights='imagenet', include_top=False, pooling='avg')
-#         self.similarity_threshold = similarity_threshold
-
-#     def load_and_preprocess_image(self, img):
-#         # Convert image to array and preprocess it
-#         img = img.convert('RGB')
-#         img_array = np.array(img)
-#         img_array = cv2.resize(img_array, (224, 224))  # Ensure correct size
-#         img_array = np.expand_dims(img_array, axis=0)
-#         img_array = preprocess_input(img_array)
-#         return img_array
-
-#     def extract_features(self, img):
-#         preprocessed_img = self.load_and_preprocess_image(img)
-#         features = self.model.predict(preprocessed_img)
-#         return features
-
-#     def calculate_ssim(self, img1, img2):
-#         img1_gray = cv2.cvtColor(img1, cv2.COLOR_RGB2GRAY)
-#         img2_gray = cv2.cvtColor(img2, cv2.COLOR_RGB2GRAY)
-#         img2_gray = cv2.resize(img2_gray, (img1_gray.shape[1], img1_gray.shape[0]))
-#         return ssim(img1_gray, img2_gray)
-
-# def process_images(reference_image, comparison_images, similarity_threshold):
-#     try:
-#         if reference_image is None:
-#             return "Please upload a reference image.", []
-#         if not comparison_images:
-#             return "Please upload comparison images.", []
-
-#         classifier = ImageCharacterClassifier(similarity_threshold)
-
-#         # Convert reference image to NumPy array
-#         ref_image = Image.fromarray(reference_image)
-#         ref_features = classifier.extract_features(ref_image)
-
-#         results = []
-#         html_output = "<h3>Comparison Results:</h3>"
-
-#         # for comp_image in comparison_images:
-#         #     try:
-#         #         # Read image file as PIL Image
-#         #         comp_pil = Image.open(comp_image)
-#         #         comp_pil = comp_pil.convert("RGB")
-
-#         #         # Convert to NumPy format for SSIM
-#         #         comp_array = np.array(comp_pil)
-
-
-#         for comp_image in comparison_images:
-#             try:
-#                 with open(comp_image.name, "rb") as f:
-#                     comp_pil = Image.open(BytesIO(f.read()))
-#                     comp_pil = comp_pil.convert("RGB")
-
-#                 comp_array = np.array(comp_pil)
-
-
-#                 # Calculate SSIM score
-#                 ssim_score = classifier.calculate_ssim(reference_image, comp_array)
-
-#                 # Extract features
-#                 comp_features = classifier.extract_features(comp_pil)
-#                 max_feature_diff = np.max(np.abs(ref_features - comp_features))
-#                 is_similar = max_feature_diff < 6.0
-
-#                 status_text = "SIMILAR" if is_similar else "NOT SIMILAR"
-#                 status_color = "green" if is_similar else "red"
-
-#                 html_output += f"<p style='color:{status_color};'>{comp_image.name}: {status_text}</p>"
-#                 results.append(comp_array)
-
-#             except Exception as e:
-#                 html_output += f"<p style='color:red;'>Error processing {comp_image.name}: {str(e)}</p>"
-
-#         return html_output, results
-
-#     except Exception as e:
-#         return f"<p style='color:red;'>Error: {str(e)}</p>", []
-
-# def create_interface():
-#     with gr.Blocks() as interface:
-#         gr.Markdown("# Image Similarity Classifier")
-#         gr.Markdown("Upload a reference image and multiple comparison images.")
-
-#         with gr.Row():
-#             with gr.Column():
-#                 reference_input = gr.Image(label="Reference Image", type="numpy")
-#                 comparison_input = gr.Files(label="Comparison Images", type="file")
-#                 threshold_slider = gr.Slider(minimum=0.0, maximum=1.0, value=0.5, step=0.05, label="Similarity Threshold")
-#                 submit_button = gr.Button("Compare Images")
-
-#             with gr.Column():
-#                 output_html = gr.HTML(label="Results")
-#                 output_gallery = gr.Gallery(label="Processed Images", columns=3)
-
-#         submit_button.click(
-#             fn=process_images,
-#             inputs=[reference_input, comparison_input, threshold_slider],
-#             outputs=[output_html, output_gallery]
-#         )
-
-#     return interface
-
-# if __name__ == "__main__":
-#     interface = create_interface()
-#     interface.launch(share=True)
-
-
-
-
-
-
-
-
-
-# import os
-# os.environ["CUDA_VISIBLE_DEVICES"] = "-1"
-
-
-# import os
-# import tempfile
-# import numpy as np
-# import cv2
-# import gradio as gr
-# from tensorflow.keras.applications import ResNet50
-# from tensorflow.keras.applications.resnet50 import preprocess_input
-# from tensorflow.keras.preprocessing import image
-# from skimage.metrics import structural_similarity as ssim
-# from PIL import Image
-# from io import BytesIO
-
-# # Disable GPU for TensorFlow
-# os.environ["TF_ENABLE_ONEDNN_OPTS"] = "0"
-# os.environ["CUDA_VISIBLE_DEVICES"] = "-1"
-
-# # --- DOCUMENTATION STRINGS (English Only) ---
-
-# GUIDELINE_SETUP = """
-# ## 1. Quick Start Guide: Setup and Run Instructions
-
-# This application uses a combination of advanced feature extraction (ResNet50) and structural analysis (SSIM) to determine if comparison images are structurally and semantically similar to a reference image.
-
-# 1.  **Upload Reference:** Upload the main image you want to compare against in the 'Reference Image' box.
-# 2.  **Upload Comparisons:** Upload one or more images you want to test for similarity in the 'Comparison Images' file upload area.
-# 3.  **Set Threshold:** Adjust the 'Similarity Threshold' slider. This primarily affects the structural (SSIM) component, but the feature comparison also plays a role (currently fixed).
-# 4.  **Run:** Click the **"Compare Images"** button.
-# 5.  **Review:** Results will appear in the 'Results' panel, indicating if each comparison image is "SIMILAR" or "NOT SIMILAR".
-# """
-
-# GUIDELINE_INPUT = """
-# ## 2. Expected Inputs
-
-# | Input Field | Purpose | Requirement |
-# | :--- | :--- | :--- |
-# | **Reference Image** | The baseline image against which all others will be compared. | Must be a single image file (JPG, PNG). |
-# | **Comparison Images** | One or more images to be tested for similarity. | Must be multiple image files. Upload them using the file selector. |
-# | **Similarity Threshold** | A slider controlling the sensitivity (0.0 to 1.0) for structural similarity (SSIM). | Higher values (closer to 1.0) mean stricter similarity requirements. Default is 0.5. |
-
-# **Image Preprocessing:** All uploaded images are automatically resized to 224x224 pixels and standardized according to the requirements of the ResNet model before feature extraction.
-# """
-
-# GUIDELINE_OUTPUT = """
-# ## 3. Expected Outputs (Similarity Results)
-
-# The application provides two main outputs:
-
-# 1.  **Results (HTML Panel):**
-#     *   A list detailing the outcome for each comparison image.
-#     *   Status: **SIMILAR** (Green) or **NOT SIMILAR** (Red).
-#     *   Similarity is determined by a combined metric: Structural Similarity (SSIM) AND feature vector distance (ResNet features).
-
-# 2.  **Processed Images (Gallery):**
-#     *   A gallery displaying the input comparison images after they have been processed.
-
-# ### How Similarity is Determined:
-# The classification relies on two checks:
-# 1.  **Feature Distance:** The distance between the deep features extracted by the ResNet50 model (checking semantic content).
-# 2.  **Structural Similarity (SSIM):** A metric comparing the structural fidelity between the reference and comparison images (checking visual layout and quality).
-# An image is typically marked "SIMILAR" only if both checks suggest a close match.
-# """
-
-# # --- CLASSIFIER CLASS ---
-# class ImageCharacterClassifier:
-#     def __init__(self, similarity_threshold=0.5):
-#         # Setting include_top=False loads the ResNet50 convolutional layers
-#         self.model = ResNet50(weights='imagenet', include_top=False, pooling='avg')
-#         self.similarity_threshold = similarity_threshold
-
-#     def load_and_preprocess_image(self, img):
-#         # Convert image to array and preprocess it
-#         img = img.convert('RGB')
-#         img_array = np.array(img)
-#         img_array = cv2.resize(img_array, (224, 224))  # Ensure correct size
-#         img_array = np.expand_dims(img_array, axis=0)
-#         img_array = preprocess_input(img_array)
-#         return img_array
-
-#     def extract_features(self, img):
-#         preprocessed_img = self.load_and_preprocess_image(img)
-#         # Use predict_on_batch for potentially better memory usage
-#         features = self.model.predict(preprocessed_img, verbose=0)
-#         return features
-
-#     def calculate_ssim(self, img1, img2):
-#         # Ensure images are in numpy array format for cv2 and SSIM
-#         img1_gray = cv2.cvtColor(img1, cv2.COLOR_RGB2GRAY)
-#         img2_gray = cv2.cvtColor(img2, cv2.COLOR_RGB2GRAY)
-        
-#         # Resize comparison image to match reference image size for SSIM calculation
-#         img2_gray = cv2.resize(img2_gray, (img1_gray.shape[1], img1_gray.shape[0]))
-        
-#         # Ensure data types are consistent (usually float/uint8 works)
-#         # SSIM calculation
-#         return ssim(img1_gray, img2_gray, data_range=img1_gray.max() - img1_gray.min())
-
-# def process_images(reference_image_array, comparison_images, similarity_threshold):
-#     try:
-#         if reference_image_array is None:
-#             return "<p style='color:red;'>Please upload a reference image.</p>", []
-#         if not comparison_images:
-#             return "<p style='color:red;'>Please upload comparison images.</p>", []
-
-#         classifier = ImageCharacterClassifier(similarity_threshold)
-
-#         # 1. Process Reference Image
-#         ref_image_pil = Image.fromarray(reference_image_array).convert("RGB")
-#         ref_features = classifier.extract_features(ref_image_pil)
-        
-#         # Convert array back to RGB for SSIM comparison later
-#         ref_image_for_ssim = cv2.cvtColor(reference_image_array, cv2.COLOR_BGR2RGB)
-
-
-#         results = []
-#         html_output = "<h3>Comparison Results:</h3>"
-
-#         # 2. Process Comparison Images
-#         for comp_file in comparison_images:
-#             try:
-#                 # Open image file using PIL
-#                 with open(comp_file.name, "rb") as f:
-#                     comp_pil = Image.open(BytesIO(f.read())).convert("RGB")
-
-#                 comp_array = np.array(comp_pil)
-                
-#                 # --- Similarity Checks ---
-
-#                 # A. SSIM Check (Structural Similarity)
-#                 ssim_score = classifier.calculate_ssim(ref_image_for_ssim, comp_array)
-#                 ssim_match = ssim_score >= similarity_threshold
-
-#                 # B. Feature Check (Semantic Similarity using ResNet features)
-#                 comp_features = classifier.extract_features(comp_pil)
-                
-#                 # Using a hardcoded feature difference threshold (6.0 in original code)
-#                 max_feature_diff = np.max(np.abs(ref_features - comp_features))
-#                 feature_match = max_feature_diff < 6.0
-                
-#                 # Combined Result
-#                 is_similar = feature_match # The original logic primarily used the feature match
-                
-#                 # If you want to require both SSIM and Feature Match:
-#                 # is_similar = ssim_match and feature_match
-                
-#                 status_text = f"SIMILAR (SSIM: {ssim_score:.3f})" if is_similar else f"NOT SIMILAR (SSIM: {ssim_score:.3f})"
-#                 status_color = "green" if is_similar else "red"
-
-#                 html_output += f"<p style='color:{status_color};'>{os.path.basename(comp_file.name)}: {status_text}</p>"
-#                 results.append(comp_array) # Add the numpy array of the comparison image
-
-#             except Exception as e:
-#                 html_output += f"<p style='color:red;'>Error processing {os.path.basename(comp_file.name)}: {str(e)}</p>"
-#                 results.append(None) # Add None to keep list consistent
-
-#         return html_output, [r for r in results if r is not None]
-
-#     except Exception as e:
-#         return f"<p style='color:red;'>Critical Error: {str(e)}</p>", []
-
-# def create_interface():
-#     with gr.Blocks(title="Image Similarity Classifier") as interface:
-        
-#         gr.Markdown("# Image Similarity Classifier (ResNet + SSIM)")
-#         gr.Markdown("Tool to compare a reference image against multiple comparison images based on structural and deep feature similarity.")
-
-#         # 1. Guidelines Section
-#         with gr.Accordion("Tips & Guidelines ", open=False):
-#             gr.Markdown(GUIDELINE_SETUP)
-#             gr.Markdown("---")
-#             gr.Markdown(GUIDELINE_INPUT)
-#             gr.Markdown("---")
-#             gr.Markdown(GUIDELINE_OUTPUT)
-            
-#         gr.Markdown("---")
-
-#         # 2. Application Interface
-#         with gr.Row():
-#             with gr.Column():
-#                 gr.Markdown("## Step 1: Upload a Reference Image ")
-#                 reference_input = gr.Image(label="Reference Image", type="numpy", height=300)
-#                 gr.Markdown("## Step 2: Upload Multiple Images to Compair with Reference Image ")
-#                 comparison_input = gr.Files(label="Comparison Images", type="file")
-#                 gr.Markdown("## Step 3: Set the Confidence Score (Optional) ")
-#                 threshold_slider = gr.Slider(minimum=0.0, maximum=1.0, value=0.5, step=0.05, label="Similarity Threshold (SSIM)")
-#                 gr.Markdown("## Step 4: Click Compare Images ")
-#                 submit_button = gr.Button("Compare Images", variant="primary")
-#                 gr.Markdown("# Results ")
-#                 gr.Markdown("## Comparison Result ")
-#                 output_html = gr.HTML(label="Comparison Results")
-#                 gr.Markdown("## Processed Comparison Images")
-#                 output_gallery = gr.Gallery(label="Processed Comparison Images", columns=3)
-
-#         # 3. Event Handling
-#         submit_button.click(
-#             fn=process_images,
-#             inputs=[reference_input, comparison_input, threshold_slider],
-#             outputs=[output_html, output_gallery]
-#         )
-        
-#         # Example data setup (Requires placeholder images to exist)
-#         gr.Markdown("---")
-#         gr.Markdown("## Sample Data for Testing")
-        
-#         # Note: You would need to provide actual file paths for reference and comparison samples
-#         # Example setup demonstrating how to structure inputs for gr.Examples:
-#         example_data = [
-#             [np.zeros((100, 100, 3), dtype=np.uint8), [gr.File("sample_data/license3.jpg"), gr.File("sample_data/licence.jpeg")], 0.6], # Placeholder example
-#         ]
-        
-#         # Since examples for Files/Gallery can be complex to set up without actual files, 
-#         # we will use a simple explanation here instead of a runnable Example block.
-#         gr.Markdown("Due to the multi-file input requirement, please manually upload a reference image and several comparison images to test.")
-
-
-#     return interface
-
-# if __name__ == "__main__":
-#     interface = create_interface()
-#     # Note: Using share=True might expose the app publicly if run without authorization.
-#     interface.launch()
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
 import os
 import numpy as np
 import cv2
@@ -625,4 +244,7 @@ if __name__ == "__main__":
     
     interface = create_interface()
     interface.queue()
-    interface.launch()
+    interface.launch(
+        server_name="0.0.0.0",
+        server_port=7860
+    )

runtime.txt

@@ -1 +0,0 @@
-python-3.10.13