push to testing branch

.gitignore

@@ -60,4 +60,6 @@ models/.gitattributes  #<-- This line can stay if you only want to ignore that f
 
 todo.md
 np_text_model
-IMG_models
+IMG_Models
+notebooks
+static

README.md

@@ -1,9 +1,152 @@
+# AI-Contain-Checker
+
+A modular AI content detection system with support for **image classification**, **image edit detection**, **Nepali text classification**, and **general text classification**. Built for performance and extensibility, it is ideal for detecting AI-generated content in both visual and textual forms.
+
+
+## 🌟 Features
+
+### 🖼️ Image Classifier
+
+* **Purpose**: Classifies whether an image is AI-generated or a real-life photo.
+* **Model**: Fine-tuned **InceptionV3** CNN.
+* **Dataset**: Custom curated dataset with **\~79,950 images** for binary classification.
+* **Location**: [`features/image_classifier`](features/image_classifier)
+* **Docs**: [`docs/features/image_classifier.md`](docs/features/image_classifier.md)
+
+### 🖌️ Image Edit Detector
+
+* **Purpose**: Detects image tampering or post-processing.
+* **Techniques Used**:
+
+  * **Error Level Analysis (ELA)**: Visualizes compression artifacts.
+  * **Fast Fourier Transform (FFT)**: Detects unnatural frequency patterns.
+* **Location**: [`features/image_edit_detector`](features/image_edit_detector)
+* **Docs**:
+
+  * [ELA](docs/detector/ELA.md)
+  * [FFT](docs/detector/fft.md )
+  * [Metadata Analysis](docs/detector/meta.md)
+  * [Backend Notes](docs/detector/note-for-backend.md)
+
+### 📝 Nepali Text Classifier
+
+* **Purpose**: Determines if Nepali text content is AI-generated or written by a human.
+* **Model**: Based on `XLMRClassifier` fine-tuned on Nepali language data.
+* **Dataset**: Scraped dataset of **\~18,000** Nepali texts.
+* **Location**: [`features/nepali_text_classifier`](features/nepali_text_classifier)
+* **Docs**: [`docs/features/nepali_text_classifier.md`](docs/features/nepali_text_classifier.md)
+
+### 🌐 English Text Classifier
+
+* **Purpose**: Detects if English text is AI-generated or human-written.
+* **Pipeline**:
+
+  * Uses **GPT2 tokenizer** for input preprocessing.
+  * Custom binary classifier to differentiate between AI and human-written content.
+* **Location**: [`features/text_classifier`](features/text_classifier)
+* **Docs**: [`docs/features/text_classifier.md`](docs/features/text_classifier.md)
+
 ---
-title: Ai-Checker
-emoji: 🚀
-colorFrom: yellow
-colorTo: blue
-sdk: docker
-pinned: false
+
+## 🗂️ Project Structure
+
+```bash
+AI-Checker/
+│
+├── app.py                  # Main FastAPI entry point
+├── config.py               # Configuration settings
+├── Dockerfile              # Docker build script
+├── Procfile                # Deployment file for Heroku or similar
+├── requirements.txt        # Python dependencies
+├── README.md               # You are here 📘
+│
+├── features/               # Core detection modules
+│   ├── image_classifier/
+│   ├── image_edit_detector/
+│   ├── nepali_text_classifier/
+│   └── text_classifier/
+│
+├── docs/                   # Internal and API documentation
+│   ├── api_endpoints.md
+│   ├── deployment.md
+│   ├── detector/
+│   │   ├── ELA.md
+│   │   ├── fft.md
+│   │   ├── meta.md
+│   │   └── note-for-backend.md
+│   ├── functions.md
+│   ├── nestjs_integration.md
+│   ├── security.md
+│   ├── setup.md
+│   └── structure.md
+│
+├── IMG_Models/             # Saved image classifier model(s)
+│   └── latest-my_cnn_model.h5
+│
+├── notebooks/              # Experimental and debug notebooks
+├── static/                 # Static assets if needed
+└── test.md                 # Test notes
+````
+
 ---
 
+## 📚 Documentation Links
+
+* [API Endpoints](docs/api_endpoints.md)
+* [Deployment Guide](docs/deployment.md)
+* [Detector Documentation](docs/detector/)
+
+  * [Error Level Analysis (ELA)](docs/detector/ELA.md)
+  * [Fast Fourier Transform (FFT)](docs/detector/fft.md)
+  * [Metadata Analysis](docs/detector/meta.md)
+  * [Backend Notes](docs/detector/note-for-backend.md)
+* [Functions Overview](docs/functions.md)
+* [NestJS Integration Guide](docs/nestjs_integration.md)
+* [Security Details](docs/security.md)
+* [Setup Instructions](docs/setup.md)
+* [Project Structure](docs/structure.md)
+
+---
+
+## 🚀 Usage
+
+1. **Install dependencies**
+
+   ```bash
+   pip install -r requirements.txt
+   ```
+
+2. **Run the API**
+
+   ```bash
+   uvicorn app:app --reload
+   ```
+
+3. **Build Docker (optional)**
+
+   ```bash
+   docker build -t ai-contain-checker .
+   docker run -p 8000:8000 ai-contain-checker
+   ```
+
+---
+
+## 🔐 Security & Integration
+
+* **Token Authentication** and **IP Whitelisting** supported.
+* NestJS integration guide: [`docs/nestjs_integration.md`](docs/nestjs_integration.md)
+* Rate limiting handled using `slowapi`.
+
+---
+
+## 🛡️ Future Plans
+
+* Add **video classifier** module.
+* Expand dataset for **multilingual** AI content detection.
+* Add **fine-tuning UI** for models.
+
+---
+
+## 📄 License
+
+See full license terms here: [`LICENSE.md`](license.md)

app.py

@@ -1,44 +1,62 @@
 from fastapi import FastAPI, Request
 from slowapi import Limiter, _rate_limit_exceeded_handler
+from fastapi.responses import FileResponse
 from slowapi.middleware import SlowAPIMiddleware
 from slowapi.errors import RateLimitExceeded
 from slowapi.util import get_remote_address
 from fastapi.responses import JSONResponse
 from features.text_classifier.routes import router as text_classifier_router
-from features.nepali_text_classifier.routes import router as nepali_text_classifier_router
+from features.nepali_text_classifier.routes import (
+    router as nepali_text_classifier_router,
+)
 from features.image_classifier.routes import router as image_classifier_router
 from features.image_edit_detector.routes import router as image_edit_detector_router
+from fastapi.staticfiles import StaticFiles
 
 from config import ACCESS_RATE
 
 import requests
+
 limiter = Limiter(key_func=get_remote_address, default_limits=[ACCESS_RATE])
 
 app = FastAPI()
-
+# added the robots.txt
 # Set up SlowAPI
 app.state.limiter = limiter
-app.add_exception_handler(RateLimitExceeded, lambda request, exc: JSONResponse(
-    status_code=429,
-    content={
-        "status_code": 429,
-        "error": "Rate limit exceeded",
-        "message": "Too many requests. Chill for a bit and try again"
-    }
-))
+app.add_exception_handler(
+    RateLimitExceeded,
+    lambda request, exc: JSONResponse(
+        status_code=429,
+        content={
+            "status_code": 429,
+            "error": "Rate limit exceeded",
+            "message": "Too many requests. Chill for a bit and try again",
+        },
+    ),
+)
 app.add_middleware(SlowAPIMiddleware)
 
 # Include your routes
 app.include_router(text_classifier_router, prefix="/text")
-app.include_router(nepali_text_classifier_router,prefix="/NP")
-app.include_router(image_classifier_router,prefix="/AI-image")
-app.include_router(image_edit_detector_router,prefix="/detect")
+app.include_router(nepali_text_classifier_router, prefix="/NP")
+app.include_router(image_classifier_router, prefix="/AI-image")
+app.include_router(image_edit_detector_router, prefix="/detect")
+
 
 @app.get("/")
 @limiter.limit(ACCESS_RATE)
 async def root(request: Request):
     return {
         "message": "API is working",
-        "endpoints": ["/text/analyse", "/text/upload", "/text/analyse-sentences", "/text/analyse-sentance-file","/NP/analyse","/NP/upload","/NP/analyse-sentences","/NP/file-sentences-analyse","/AI-image/analyse"]
+        "endpoints": [
+            "/text/analyse",
+            "/text/upload",
+            "/text/analyse-sentences",
+            "/text/analyse-sentance-file",
+            "/NP/analyse",
+            "/NP/upload",
+            "/NP/analyse-sentences",
+            "/NP/file-sentences-analyse",
+            "/AI-image/analyse",
+        ],
     }
-

docs/api_endpoints.md

@@ -2,13 +2,13 @@
 
 ### English (GPT-2) - `/text/`
 
-| Endpoint                         | Method | Description                               |
-| --------------------------------- | ------ | ----------------------------------------- |
-| `/text/analyse`                  | POST   | Classify raw English text                 |
-| `/text/analyse-sentences`        | POST   | Sentence-by-sentence breakdown            |
-| `/text/analyse-sentance-file`    | POST   | Upload file, per-sentence breakdown       |
-| `/text/upload`                   | POST   | Upload file for overall classification    |
-| `/text/health`                   | GET    | Health check                             |
+| Endpoint                      | Method | Description                            |
+| ----------------------------- | ------ | -------------------------------------- |
+| `/text/analyse`               | POST   | Classify raw English text              |
+| `/text/analyse-sentences`     | POST   | Sentence-by-sentence breakdown         |
+| `/text/analyse-sentance-file` | POST   | Upload file, per-sentence breakdown    |
+| `/text/upload`                | POST   | Upload file for overall classification |
+| `/text/health`                | GET    | Health check                           |
 
 #### Example: Classify English text
 
@@ -20,6 +20,7 @@ curl -X POST http://localhost:8000/text/analyse \
 ```
 
 **Response:**
+
 ```json
 {
   "result": "AI-generated",
@@ -40,13 +41,13 @@ curl -X POST http://localhost:8000/text/upload \
 
 ### Nepali (SentencePiece) - `/NP/`
 
-| Endpoint                         | Method | Description                               |
-| --------------------------------- | ------ | ----------------------------------------- |
-| `/NP/analyse`                    | POST   | Classify Nepali text                      |
-| `/NP/analyse-sentences`          | POST   | Sentence-by-sentence breakdown            |
-| `/NP/upload`                     | POST   | Upload Nepali PDF for classification      |
-| `/NP/file-sentences-analyse`     | POST   | PDF upload, per-sentence breakdown        |
-| `/NP/health`                     | GET    | Health check                             |
+| Endpoint                     | Method | Description                          |
+| ---------------------------- | ------ | ------------------------------------ |
+| `/NP/analyse`                | POST   | Classify Nepali text                 |
+| `/NP/analyse-sentences`      | POST   | Sentence-by-sentence breakdown       |
+| `/NP/upload`                 | POST   | Upload Nepali PDF for classification |
+| `/NP/file-sentences-analyse` | POST   | PDF upload, per-sentence breakdown   |
+| `/NP/health`                 | GET    | Health check                         |
 
 #### Example: Nepali text classification
 
@@ -58,6 +59,7 @@ curl -X POST http://localhost:8000/NP/analyse \
 ```
 
 **Response:**
+
 ```json
 {
   "label": "Human",
@@ -73,20 +75,18 @@ curl -X POST http://localhost:8000/NP/upload \
   -F 'file=@NepaliText.pdf;type=application/pdf'
 ```
 
-
 ### Image-Classification -`/verify-image/`
- 
-| Endpoint                         | Method | Description                               |
-| --------------------------------- | ------ | ----------------------------------------- |
-| `/verify-image/analyse`           | POST   | Classify Image using ML                   |
 
+| Endpoint                | Method | Description             |
+| ----------------------- | ------ | ----------------------- |
+| `/verify-image/analyse` | POST   | Classify Image using ML |
+
+#### Example: Image-Classification
 
-#### Example: Image-Classification 
 ```bash
 curl -X POST http://localhost:8000/verify-image/analyse \
   -H "Authorization: Bearer <SECRET_TOKEN>" \
   -F 'file=@test1.png'
 ```
 
-
-
+[🔙 Back to Main README](../README.md)

docs/deployment.md

@@ -105,3 +105,4 @@ Happy deploying!
 **P.S.** Try not to break stuff. 😅
 
 
+[🔙 Back to Main README](../README.md)

docs/detector/ELA.md

@@ -2,14 +2,12 @@
 
 This module provides a function to perform Error Level Analysis (ELA) on images to detect potential manipulations or edits.
 
-
 ## Function: `run_ela`
 
 ```python
 def run_ela(image: Image.Image, quality: int = 90, threshold: int = 15) -> bool:
 ```
 
-
 ### Description
 
 Error Level Analysis (ELA) works by recompressing an image at a specified JPEG quality level and comparing it to the original image. Differences between the two images reveal areas with inconsistent compression artifacts — often indicating image manipulation.
@@ -24,14 +22,12 @@ The function computes the maximum pixel difference across all color channels and
 | `quality`   | `int`       | 90      | JPEG compression quality used for recompression during analysis (lower = more compression). |
 | `threshold` | `int`       | 15      | Pixel difference threshold to flag the image as edited.                                     |
 
-
 ### Returns
 
 `bool`
 
-* `True` if the image is likely edited (max pixel difference > threshold).  
-* `False` if the image appears unedited.
-
+- `True` if the image is likely edited (max pixel difference > threshold).
+- `False` if the image appears unedited.
 
 ### Usage Example
 
@@ -48,13 +44,11 @@ is_edited = run_ela(img, quality=90, threshold=15)
 print("Image edited:", is_edited)
 ```
 
-
 ### Notes
 
-* The input image **must** be in RGB mode for accurate analysis.  
-* ELA is a heuristic technique; combining it with other detection methods increases reliability.  
-* Visualizing the enhanced difference image can help identify edited regions (not returned by this function but possible to add).
-
+- The input image **must** be in RGB mode for accurate analysis.
+- ELA is a heuristic technique; combining it with other detection methods increases reliability.
+- Visualizing the enhanced difference image can help identify edited regions (not returned by this function but possible to add).
 
 ### Installation
 
@@ -64,13 +58,8 @@ Make sure you have Pillow installed:
 pip install pillow
 ```
 
-
 ### Running Locally
 
 Just put the function in a notebook or script file and run it with your image. It works well for basic images.
 
-
-### Developer
-
-Pujan Neupane
-
+[🔙 Back to Main README](../README.md)

docs/detector/fft.md

@@ -133,9 +133,4 @@ it is implemented in the api
 Just put the function in a notebook or script file and run it with your image. It works well for basic images.
 
 
-### Worked By
-
-Pujan Neupane
-
-
-
+[🔙 Back to Main README](../README.md)

docs/detector/meta.md

@@ -0,0 +1,20 @@
+# Metadata Analysis for Image Edit Detection
+
+This module inspects image metadata to detect possible signs of AI-generation or post-processing edits.
+
+## Overview
+
+- Many AI-generated images and edited images leave identifiable traces in their metadata.
+- This detector scans image EXIF metadata and raw bytes for known AI generation indicators and common photo editing software signatures.
+- It classifies images as `"ai_generated"`, `"edited"`, or `"undetermined"` based on detected markers.
+- Handles invalid image formats gracefully by reporting errors.
+
+## How It Works
+
+- Opens the image from raw bytes using the Python Pillow library (`PIL`).
+- Reads EXIF metadata and specifically looks for the "Software" tag that often contains the editing app name.
+- Checks for common image editors such as Photoshop, GIMP, Snapseed, etc.
+- Scans the entire raw byte content of the image for embedded AI generation identifiers like "midjourney", "stable-diffusion", "openai", etc.
+- Returns a status string indicating the metadata classification.
+
+[🔙 Back to Main README](../README.md)

docs/detector/note-for-backend.md

@@ -90,3 +90,5 @@ POST /api/detect-image
 | Metadata | 🚀 Fast     | ⚠️ Low confidence    |
 
 > For high-throughput systems, consider running Metadata first and conditionally applying ELA/FFT if suspicious.
+
+[🔙 Back to Main README](../README.md)

docs/features/image_classifier.md

@@ -0,0 +1,31 @@
+# Image Classifier
+
+## Overview
+
+This module classifies whether an input image is AI-generated or a real-life photograph.
+
+## Model
+
+- Architecture: InceptionV3
+- Type: Binary Classifier (AI vs Real)
+- Format: H5 model (`latest-my_cnn_model.h5`)
+
+## Dataset
+
+- Total images: ~79,950
+- Balanced between real and generated images
+- Preprocessing: Resizing, normalization
+
+## Code Location
+
+- Controller: `features/image_classifier/controller.py`
+- Model Loader: `features/image_classifier/model_loader.py`
+- Preprocessor: `features/image_classifier/preprocess.py`
+
+## API
+
+- Endpoint: [ENDPOINTS](../api_endpoints.md)
+- Input: Image file (PNG/JPG)
+- Output: JSON response with classification result and confidence
+
+[🔙 Back to Main README](../README.md)

docs/features/nepali_text_classifier.md

@@ -0,0 +1,30 @@
+# Nepali Text Classifier
+
+## Overview
+
+This classifier identifies whether Nepali-language text content is written by a human or AI.
+
+## Model
+
+- Base Model: XLM-Roberta (XLMRClassifier)
+- Language: Nepali (Multilingual model)
+- Fine-tuned with scraped web content (~18,000 samples)
+
+## Dataset
+
+- Custom scraped dataset with manual labeling
+- Includes news, blogs, and synthetic content from various LLMs
+
+## Code Location
+
+- Controller: `features/nepali_text_classifier/controller.py`
+- Inference: `features/nepali_text_classifier/inferencer.py`
+- Model Loader: `features/nepali_text_classifier/model_loader.py`
+
+## API
+
+- Endpoint: [ENDPOINTS](../api_endpoints.md)
+- Input: Raw text
+- Output: JSON classification with label and confidence score
+
+[🔙 Back to Main README](../README.md)

docs/features/text_classifier.md

@@ -0,0 +1,30 @@
+# English Text Classifier
+
+## Overview
+
+Detects whether English-language text is AI-generated or human-written.
+
+## Model Pipeline
+
+- Tokenizer: GPT-2 Tokenizer
+- Model: Custom trained binary classifier
+
+## Dataset
+
+- Balanced dataset: Human vs AI-generated (ChatGPT, Claude, etc.)
+- Tokenized and fed into the model using PyTorch/TensorFlow
+
+## Code Location
+
+- Controller: `features/text_classifier/controller.py`
+- Inference: `features/text_classifier/inferencer.py`
+- Model Loader: `features/text_classifier/model_loader.py`
+- Preprocessor: `features/text_classifier/preprocess.py`
+
+## API
+
+- Endpoint: [ENDPOINTS](../api_endpoints.md)
+- Input: Raw English text
+- Output: Prediction result with probability/confidence
+
+[🔙 Back to Main README](../README.md)

docs/functions.md

@@ -52,12 +52,11 @@
 ---
 ## for image_classifier
 
-
-
 - **`Classify_Image_router()`** – Handles image classification requests by routing and coordinating preprocessing and inference.
 - **`classify_image()`** – Performs AI vs human image classification using the loaded model.
 - **`load_model()`** – Loads the pretrained model from Hugging Face at server startup.
 - **`preprocess_image()`** – Applies all required preprocessing steps to the input image.
 
+> Note: While many functions mirror those in the text classifier, the image classifier primarily uses TensorFlow rather than PyTorch.
 
-
+[🔙 Back to Main README](../README.md)

docs/nestjs_integration.md

@@ -80,3 +80,4 @@ export class AppController {
   }
 }
 ```
+[🔙 Back to Main README](../README.md)

docs/security.md

@@ -7,3 +7,4 @@ All endpoints require authentication via Bearer token:
 
 Unauthorized requests receive `403 Forbidden`.
 
+[🔙 Back to Main README](../README.md)

docs/setup.md

@@ -21,3 +21,4 @@ SECRET_TOKEN=your_secret_token_here
 ```bash
 uvicorn app:app --host 0.0.0.0 --port 8000
 ```
+[🔙 Back to Main README](../README.md)

docs/status_code.md

@@ -0,0 +1,68 @@
+# Error Codes Reference
+
+## 🔹 Summary Table
+
+| Code | Message                                               | Description                                |
+| ---- | ----------------------------------------------------- | ------------------------------------------ |
+| 400  | Text must contain at least two words                  | Input text too short                       |
+| 400  | Text should be less than 10,000 characters            | Input text too long                        |
+| 404  | The file is empty or only contains whitespace         | File has no usable content                 |
+| 404  | Invalid file type. Only .docx, .pdf, and .txt allowed | Unsupported file format                    |
+| 403  | Invalid or expired token                              | Authentication token is invalid or expired |
+| 413  | Text must contain at least two words                  | Text too short (alternative condition)     |
+| 413  | Text must be less than 10,000 characters              | Text too long (alternative condition)      |
+| 413  | The image error (preprocessing)                       | Image size/content issue                   |
+| 500  | Error processing the file                             | Internal server error while processing     |
+
+---
+
+## 🔍 Error Details
+
+### `400` - Bad Request
+
+- **Text must contain at least two words**  
+  The input text field is too short. Submit at least two words to proceed.
+
+- **Text should be less than 10,000 characters**  
+  Input text exceeds the maximum allowed character limit. Consider truncating or summarizing the content.
+
+---
+
+### `404` - Not Found
+
+- **The file is empty or only contains whitespace**  
+  The uploaded file is invalid due to lack of meaningful content. Ensure the file has readable, non-empty text.
+
+- **Invalid file type. Only .docx, .pdf, and .txt are allowed**  
+  The file format is not supported. Convert the file to one of the allowed formats before uploading.
+
+---
+
+### `403` - Forbidden
+
+- **Invalid or expired token**  
+  Your access token is either expired or incorrect. Try logging in again or refreshing the token.
+
+---
+
+### `413` - Payload Too Large
+
+- **Text must contain at least two words**  
+  The text payload is too small or malformed under a large upload context. Add more content.
+
+- **Text must be less than 10,000 characters**  
+  The payload exceeds the allowed character limit for a single request. Break it into smaller chunks if needed.
+
+- **The image error**  
+  The uploaded image is too large or corrupted. Try resizing or compressing it before retrying.
+
+---
+
+### `500` - Internal Server Error
+
+- **Error processing the file**  
+  An unexpected server-side failure occurred during file analysis. Retry later or contact support if persistent.
+
+---
+
+> 📌 **Note:** Always validate inputs, check token status, and follow file guidelines before making requests.

docs/structure.md

@@ -1,36 +1,58 @@
 ## 🏗️ Project Structure
 
-```
-├── app.py                   # Main FastAPI app entrypoint
-├── config.py                # Configuration loader (.env, settings)
-├── features/
-│   ├── text_classifier/     # English (GPT-2) classifier
+```bash
+AI-Checker/
+│
+├── app.py                  # Main FastAPI entry point
+├── config.py               # Configuration settings
+├── Dockerfile              # Docker build script
+├── Procfile                # Deployment entry for platforms like Heroku/Railway
+├── requirements.txt        # Python dependency list
+├── README.md               # Main project overview 📘
+│
+├── features/               # Core AI content detection modules
+│   ├── image_classifier/           # Classifies AI vs Real images
+│   │   ├── controller.py
+│   │   ├── model_loader.py
+│   │   └── preprocess.py
+│   ├── image_edit_detector/       # Detects tampered or edited images
+│   ├── nepali_text_classifier/    # Classifies Nepali text as AI or Human
 │   │   ├── controller.py
 │   │   ├── inferencer.py
 │   │   ├── model_loader.py
-│   │   ├── preprocess.py
-│   │   └── routes.py
-│   └── nepali_text_classifier/ # Nepali (sentencepiece) classifier
+│   │   └── preprocess.py
+│   └── text_classifier/           # Classifies English text as AI or Human
 │       ├── controller.py
 │       ├── inferencer.py
 │       ├── model_loader.py
-│       ├── preprocess.py
-│       └── routes.py
-├── np_text_model/           # Nepali model artifacts (auto-downloaded)
-│   ├── classifier/
-│   │   └── sentencepiece.bpe.model
-│   └── model_95_acc.pth
-├── models/                  # English GPT-2 model/tokenizer (auto-downloaded)
-│   ├── merges.txt
-│   ├── tokenizer.json
-│   └── model_weights.pth
-├── Dockerfile               # Container build config
-├── Procfile                 # Deployment entrypoint (for PaaS)
-├── requirements.txt         # Python dependencies
-├── README.md     
-├── Docs                     # documents
-└── .env                     # Secret token(s), environment config
+│       └── preprocess.py
+│
+├── docs/                   # Internal documentation and API references
+│   ├── api_endpoints.md
+│   ├── deployment.md
+│   ├── detector/
+│   │   ├── ELA.md
+│   │   ├── fft.md
+│   │   ├── meta.md
+│   │   └── note-for-backend.md
+│   ├── features/
+│   │   ├── image_classifier.md
+│   │   ├── nepali_text_classifier.md
+│   │   └── text_classifier.md
+│   ├── functions.md
+│   ├── nestjs_integration.md
+│   ├── security.md
+│   ├── setup.md
+│   └── structure.md
+│
+├── IMG_Models/             # Stored model weights
+│   └── latest-my_cnn_model.h5
+│
+├── notebooks/              # Experimental/debug Jupyter notebooks
+├── static/                 # Static files (e.g., UI assets, test inputs)
+└── test.md                 # Test usage notes
 ```
+
 ### 🌟 Key Files and Their Roles
 
 - **`app.py`**: Entry point initializing FastAPI app and routes.
@@ -39,16 +61,14 @@
 - **`__init__.py`**: Package initializer for the root module and submodules.
 - **`features/text_classifier/`**
   - **`controller.py`**: Handles logic between routes and the model.
-  - **`inferencer.py`**: Runs inference and returns predictions as well as file system 
-  utilities.
+  - **`inferencer.py`**: Runs inference and returns predictions as well as file system
+    utilities.
 - **`features/NP/`**
   - **`controller.py`**: Handles logic between routes and the model.
-  - **`inferencer.py`**: Runs inference and returns predictions as well as file system 
-  utilities.
+  - **`inferencer.py`**: Runs inference and returns predictions as well as file system
+    utilities.
   - **`model_loader.py`**: Loads the ML model and tokenizer.
   - **`preprocess.py`**: Prepares input text for the model.
   - **`routes.py`**: Defines API routes for text classification.
 
-
-
--[Main](../README.md)
+[🔙 Back to Main README](../README.md)

features/image_classifier/controller.py

@@ -1,11 +1,16 @@
-from fastapi import HTTPException,File,UploadFile
+from fastapi import HTTPException, File, UploadFile
 from .preprocess import preprocess_image
 from .inferencer import classify_image
+
+
 async def Classify_Image_router(file: UploadFile = File(...)):
     try:
         image_array = preprocess_image(file)
-        result = classify_image(image_array)
-        return result
-    except Exception as e:
-        raise HTTPException(status_code=400, detail=str(e))
+        try:
+            result = classify_image(image_array)
+            return result
+        except:
+            raise HTTPException(status_code=423, detail="something went wrong")
 
+    except Exception as e:
+        raise HTTPException(status_code=413, detail=str(e))

features/image_classifier/inferencer.py

@@ -1,22 +1,42 @@
 import numpy as np
-from .model_loader import load_model
-model = load_model()
-
-def classify_image(image: np.ndarray):
-    predictions = model.predict(image)[0]
-    human_conf = float(predictions[0])
-    ai_conf = float(predictions[1])
-
-    if ai_conf > 0.55:
-        label = "AI Generated"
-    elif ai_conf < 0.45:
-        label = "Human Generated"
-    else:
-        label = "Maybe AI"
-
-    return {
-        "label": label,
-        "ai_confidence": round(ai_conf * 100, 2),
-        "human_confidence": round(human_conf * 100, 2)
-    }
+from .model_loader import get_model
 
+# Thresholds
+AI_THRESHOLD = 0.55
+HUMAN_THRESHOLD = 0.45
+
+
+def classify_image(image_array: np.ndarray) -> dict:
+    try:
+        model = get_model()
+        predictions = model.predict(image_array)
+
+        if predictions.ndim != 2 or predictions.shape[1] != 1:
+            raise ValueError(
+                "Model output shape is invalid. Expected shape: (batch, 1)"
+            )
+
+        ai_conf = float(np.clip(predictions[0][0], 0.0, 1.0))
+        human_conf = 1.0 - ai_conf
+
+        # Classification logic
+        if ai_conf > AI_THRESHOLD:
+            label = "AI Generated"
+        elif ai_conf < HUMAN_THRESHOLD:
+            label = "Human Generated"
+        else:
+            label = "Uncertain (Maybe AI)"
+
+        return {
+            "label": label,
+            "ai_confidence": round(ai_conf * 100, 2),
+            "human_confidence": round(human_conf * 100, 2),
+        }
+
+    except Exception as e:
+        return {
+            "error": str(e),
+            "label": "Classification Failed",
+            "ai_confidence": None,
+            "human_confidence": None,
+        }

features/image_classifier/model_loader.py

@@ -1,43 +1,58 @@
-import tensorflow as tf
-from tensorflow.keras.models import load_model as keras_load_model
 import os
-from huggingface_hub import snapshot_download
 import shutil
+import logging
+import tensorflow as tf
+from tensorflow.keras.layers import Layer
+from huggingface_hub import snapshot_download
 
-# Constants
+# Model config
 REPO_ID = "can-org/AI-VS-HUMAN-IMAGE-classifier"
-MODEL_DIR = "./IMG_models"
-MODEL_PATH = os.path.join(MODEL_DIR, 'latest-my_cnn_model.h5')  # adjust path as needed
+MODEL_DIR = "./IMG_Models"
+WEIGHTS_PATH = os.path.join(MODEL_DIR, "latest-my_cnn_model.h5")
+
+# Device info (for logging)
+gpus = tf.config.list_physical_devices("GPU")
+device = "cuda" if gpus else "cpu"
+
+# Global model reference
+_model_img = None
 
-_model_img = None  # global model variable
+# Custom layer used in the model
+class Cast(Layer):
+    def call(self, inputs):
+        return tf.cast(inputs, tf.float32)
 
 def warmup():
     global _model_img
-    if not os.path.exists(MODEL_DIR):
-        download_model_Repo()
+    download_model_repo()
     _model_img = load_model()
+    logging.info("Image model is ready.")
 
-def download_model_Repo():
-    if os.path.exists(MODEL_DIR):
+def download_model_repo():
+    if os.path.exists(MODEL_DIR) and os.path.isdir(MODEL_DIR):
+        logging.info("Image model already exists, skipping download.")
         return
     snapshot_path = snapshot_download(repo_id=REPO_ID)
     os.makedirs(MODEL_DIR, exist_ok=True)
     shutil.copytree(snapshot_path, MODEL_DIR, dirs_exist_ok=True)
 
 def load_model():
-    if not os.path.exists(MODEL_DIR):
-        download_model_Repo()
-
-    # Check for GPU availability
-    gpus = tf.config.list_physical_devices('GPU')
-    if gpus:
-        # GPU is available, load model normally
-        print("GPU detected, loading model on GPU.")
-        model = keras_load_model(MODEL_PATH)
-    else:
-        # No GPU, force CPU usage
-        print("No GPU detected, forcing model loading on CPU.")
-        with tf.device('/CPU:0'):
-            model = keras_load_model(MODEL_PATH)
-    return model
+    global _model_img
+    if _model_img is not None:
+        return _model_img
+
+    print(f"{'GPU detected' if device == 'cuda' else 'No GPU detected'}, loading model on {device.upper()}.")
+
+    _model_img = tf.keras.models.load_model(
+        WEIGHTS_PATH, custom_objects={"Cast": Cast}
+    )
+    print("Model input shape:", _model_img.input_shape)
+    return _model_img
+
+def get_model():
+    global _model_img
+    if _model_img is None:
+        download_model_repo()
+        _model_img = load_model()
+    return _model_img
 

features/image_classifier/preprocess.py

@@ -1,18 +1,26 @@
 import numpy as np
 import cv2
+from fastapi import HTTPException
+
 
 def preprocess_image(file):
-    # Read bytes from UploadFile
-    image_bytes = file.file.read()
-    # Convert bytes to NumPy array
-    nparr = np.frombuffer(image_bytes, np.uint8)
-    img = cv2.imdecode(nparr, cv2.IMREAD_COLOR)
-    if img is None:
-        raise ValueError("Could not decode image.")
+    try:
+        file.file.seek(0)
+        image_bytes = file.file.read()
+        nparr = np.frombuffer(image_bytes, np.uint8)
+        img = cv2.imdecode(nparr, cv2.IMREAD_COLOR)
+        if img is None:
+            raise HTTPException(status_code=500, detail="Could not decode image.")
 
-    img = cv2.resize(img, (256, 256))  # Changed size to 256x256
-    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
-    img = img / 255.0
-    img = np.expand_dims(img, axis=0)
-    return img
+        img = cv2.resize(img, (299, 299))
+        img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
+        img = img / 255.0
+        img = np.expand_dims(img, axis=0).astype(np.float32)
+        return img
 
+    except HTTPException:
+        raise  # Re-raise already defined HTTP errors
+    except Exception as e:
+        raise HTTPException(
+            status_code=500, detail=f"Image preprocessing failed: {str(e)}"
+        )

features/image_edit_detector/detectors/ela.py

@@ -1,6 +1,7 @@
 from PIL import Image, ImageChops, ImageEnhance
 import io
 
+
 def run_ela(image: Image.Image, quality: int = 90, threshold: int = 15) -> bool:
     """
     Perform Error Level Analysis to detect image manipulation.
@@ -16,7 +17,7 @@ def run_ela(image: Image.Image, quality: int = 90, threshold: int = 15) -> bool:
 
     # Recompress the image into JPEG format in memory
     buffer = io.BytesIO()
-    image.save(buffer, format='JPEG', quality=quality)
+    image.save(buffer, format="JPEG", quality=quality)
     buffer.seek(0)
     recompressed = Image.open(buffer)
 
@@ -29,4 +30,3 @@ def run_ela(image: Image.Image, quality: int = 90, threshold: int = 15) -> bool:
     _ = ImageEnhance.Brightness(diff).enhance(10)
 
     return max_diff > threshold
-

license.md

@@ -0,0 +1,20 @@
+# License - All Rights Reserved
+
+Copyright (c) 2025 CyberAlertNepal
+
+This software and all associated materials are **not open source** and are protected under a custom license.
+
+## Strict Usage Terms
+
+Unless explicit written permission is granted by **CyberAlertNepal**, **no individual or entity** is allowed to:
+
+- Use this codebase or its models in any capacity — personal, educational, or commercial.
+- Modify, copy, distribute, or sublicense any part of this project.
+- Deploy, mirror, or host this project, either publicly or privately.
+- Incorporate any component of this project into derivative works or other applications.
+
+This project is intended for **private, internal use by the author(s) only**.
+
+Any unauthorized usage, reproduction, or distribution is strictly prohibited and may result in legal action.
+
+**All rights reserved.**

requirements.txt

@@ -15,3 +15,4 @@ tensorflow
 opencv-python
 pillow
 scipy
+fitz