app.py

import gradio as gr
from ultralytics import YOLO
from PIL import Image, ImageDraw
import pytesseract
import subprocess

# Ensure Tesseract OCR is installed and detected
TESSERACT_PATH = "/usr/bin/tesseract"
pytesseract.pytesseract.tesseract_cmd = TESSERACT_PATH

def check_tesseract():
    """Check if Tesseract is installed and print its version."""
    try:
        tesseract_version = subprocess.check_output([TESSERACT_PATH, "--version"]).decode("utf-8").split("\n")[0]
        print(f"Tesseract Version: {tesseract_version}")
        return True
    except Exception as e:
        print(f"Tesseract not found: {e}")
        return False

# Load YOLO model (ensure best.pt exists in the working directory)
YOLO_MODEL_PATH = "best.pt"
model = YOLO(YOLO_MODEL_PATH, task='detect').to("cpu")

def merge_boxes_into_lines(boxes, y_threshold=10):
    """Merge bounding boxes if they belong to the same text row."""
    if len(boxes) == 0:
        return []

    boxes = sorted(boxes, key=lambda b: b[1])  # Sort by y-axis (top position)
    merged_lines = []
    current_line = list(boxes[0])

    for i in range(1, len(boxes)):
        x1, y1, x2, y2 = boxes[i]

        if abs(y1 - current_line[1]) < y_threshold:  # Close enough to the previous line
            current_line[0] = min(current_line[0], x1)  # Extend left boundary
            current_line[2] = max(current_line[2], x2)  # Extend right boundary
            current_line[3] = max(current_line[3], y2)  # Extend bottom boundary
        else:
            merged_lines.append(current_line)
            current_line = list(boxes[i])

    merged_lines.append(current_line)
    return merged_lines

def detect_and_ocr(image):
    """Detects text lines, draws bounding boxes, and runs OCR if available."""
    image = Image.fromarray(image)
    original_image = image.copy()

    results = model.predict(image, conf=0.3, iou=0.5, device="cpu")
    detected_boxes = results[0].boxes.xyxy.tolist()
    detected_boxes = [list(map(int, box)) for box in detected_boxes]

    merged_boxes = merge_boxes_into_lines(detected_boxes)

    draw = ImageDraw.Draw(original_image)
    extracted_text_lines = []

    for idx, (x1, y1, x2, y2) in enumerate(merged_boxes):
        draw.rectangle([x1, y1, x2, y2], outline="blue", width=2)
        draw.text((x1, y1 - 10), f"Line {idx}", fill="blue")

        cropped_line = image.crop((x1, y1, x2, y2))

        if check_tesseract():  # If Tesseract is installed, run OCR
            try:
                ocr_text = pytesseract.image_to_string(cropped_line, lang="khm+eng").strip()
                if ocr_text:
                    extracted_text_lines.append(ocr_text)
            except Exception as e:
                print(f"OCR failed for line {idx}: {e}")

    full_text = "\n".join(extracted_text_lines) if extracted_text_lines else "⚠️ OCR not available. Showing detected lines only."

    return original_image, full_text

# Gradio UI
with gr.Blocks() as iface:
    gr.Markdown("# 📜 Text Line Detection with Khmer OCR")
    gr.Markdown("## 📷 Upload an image to detect text lines and extract Khmer text")

    with gr.Row():
        with gr.Column(scale=1):
            gr.Markdown("### 📤 Upload Image")
            image_input = gr.Image(type="numpy", label="Upload an image")

        with gr.Column(scale=1):
            gr.Markdown("### 🖼 Annotated Image with Bounding Boxes")
            output_annotated = gr.Image(type="pil", label="Detected Text Lines")

    gr.Markdown("### 📝 Extracted Text (OCR Result)")
    output_text = gr.Textbox(label="Extracted Text", lines=10)

    image_input.upload(
        detect_and_ocr,
        inputs=image_input,
        outputs=[output_annotated, output_text]
    )

# 🚀 Ensure the app runs properly in Hugging Face Spaces
if __name__ == "__main__":
    iface.launch(server_name="0.0.0.0", server_port=7860)