app.py

import gradio as gr
from PIL import Image
from ultralytics import YOLO
import requests
import json
import logging

logging.basicConfig(level=logging.INFO)

model = YOLO("BP_Multiple_Objects_Complicated_v1.pt")

def detect_objects(images):
    results = model(images)
    all_bboxes = []
    all_bboxes2 = []
    all_segments = []
    for result in results:
        boxes = result.boxes.xywhn.tolist()
        boxes2 = result.boxes.xywh.tolist()
        all_bboxes.append(boxes)
        all_bboxes2.append(boxes2)

        if result.masks is not None:
            masks = result.masks.xyn                         
            sub_arrays = [arr.tolist() for arr in masks]
        else:
            sub_arrays = []
            
        all_segments.append(sub_arrays)

    return all_bboxes, all_bboxes2, all_segments

def create_solutions(image_urls, all_bboxes, all_bboxes2, all_segments):
    solutions = []   
    img_id = 1
    box_id = 1
    cat_id = 1
    for image_url, bbox, bbox2, segmnt in zip(image_urls, all_bboxes, all_bboxes2, all_segments):       

        for subbox, subbox2, subsegmnt in zip(bbox, bbox2, segmnt):
            w = subbox2[2]
            h = subbox2[3]
            area = w * h

            flattened_segmnt = [item for sublist in subsegmnt for item in sublist]

            ans = {"image_id": img_id, "id": box_id, "area": area, "category_id": cat_id, "bbox": subbox, "segment": flattened_segmnt}
            obj ={"url": image_url, "answer":[ans]}
            solutions.append(obj)
            box_id += 1
        img_id += 1
    return solutions


# def send_results_to_api(data, result_url):
#     headers = {"Content-Type": "application/json"}
#     response = requests.post(result_url, json=data, headers=headers)
#     if response.status_code == 200:
#         return response.json()
#     else:
#         return {"error": f"Failed to send results to API: {response.status_code}"}


def process_images(params):
    try:
        params = json.loads(params)
    except json.JSONDecodeError as e:
        logging.error(f"Invalid JSON input: {e.msg} at line {e.lineno} column {e.colno}")
        return {"error": f"Invalid JSON input: {e.msg} at line {e.lineno} column {e.colno}"}
    
    image_urls = params.get("urls", [])
    # api = params.get("api", "")
    # job_id = params.get("job_id", "")

    if not image_urls:
        logging.error("Missing required parameters: 'urls'")
        return {"error": "Missing required parameters: 'urls'"}

    try:
        images = [Image.open(requests.get(url, stream=True).raw) for url in image_urls]
    except Exception as e:
        logging.error(f"Error loading images: {e}")
        return {"error": f"Error loading images: {str(e)}"}
    
    all_bboxes, all_bboxes2, all_segments = detect_objects(images)
    solutions = create_solutions(image_urls, all_bboxes, all_bboxes2, all_segments)

    # result_url = f"{api}/{job_id}"
    # send_results_to_api(solutions, result_url)

    return json.dumps({"solutions": solutions})


inputt = gr.Textbox(label="Parameters (JSON format)")
outputs = gr.JSON()

application = gr.Interface(fn=process_images, inputs=inputt, outputs=outputs, title="Multiple Object Segmentation with API Integration")
application.launch()





# import gradio as gr
# from PIL import Image
# from ultralytics import YOLO
# import requests
# import json

# model = YOLO("BP_Multiple_Objects_Complicated_v1.pt")

# def detect_objects(images):
#     results = model(images)
#     all_bboxes = []
#     all_bboxes2 = []
#     all_segments = []
#     for result in results:
#         boxes = result.boxes.xywhn.tolist()
#         boxes2 = result.boxes.xywh.tolist()
#         all_bboxes.append(boxes)
#         all_bboxes2.append(boxes2)

#         masks = result.masks.xyn                         
#         sub_arrays = [arr.tolist() for arr in masks]     
#         all_segments.append(sub_arrays)

#         return all_bboxes, all_bboxes2, all_segments

# def create_solutions(image_urls, all_bboxes, all_bboxes2, all_segments):
#     solutions = []   
#     img_id =1
#     box_id =1
#     cat_id =1
#     for image_url, bbox, bbox2, segmnt in zip(image_urls, all_bboxes, all_bboxes2, all_segments):       

#         for subbox, subbox2, subsegmnt in zip(bbox, bbox2, segmnt):

#             w = subbox2[2]
#             h = subbox2[3]
#             area = w*h

#             flattened_segmnt = [item for sublist in subsegmnt for item in sublist]

#             obj = {"image_id":img_id, "image_url": image_url, "id":box_id, "area":area, "category_id":cat_id, "bbox": subbox, "segment":flattened_segmnt}                 # Create an object for each image
#             box_id +=1
#             solutions.append(obj)                     
#         img_id +=1
#     return solutions

# def send_results_to_api(data, result_url):
#     # Example function to send results to an API
#     headers = {"Content-Type": "application/json"}
#     response = requests.post(result_url, json=data, headers=headers)
#     if response.status_code == 200:
#         return response.json()  # Return any response from the API if needed
#     else:
#         return {"error": f"Failed to send results to API: {response.status_code}"}

# def process_images(params):
#     # Parse the JSON string into a dictionary
#     params = json.loads(params)
    
#     image_urls = params.get("image_urls", [])
#     api = params.get("api", "")
#     job_id = params.get("job_id", "")

#     images = [Image.open(requests.get(url, stream=True).raw) for url in image_urls]  # images from URLs

#     all_bboxes, all_bboxes2, all_segments = detect_objects(images)  # Perform object detection
#     solutions = create_solutions(image_urls, all_bboxes, all_bboxes2, all_segments)  # Create solutions with image URLs and bounding boxes

#     result_url = f"{api}/{job_id}"
#     # send_results_to_api(solutions, result_url)

#     return json.dumps({"solutions": solutions}, indent=4)


# inputt = gr.Textbox(label="Parameters (JSON format)")
# outputs = gr.JSON()

# application = gr.Interface(fn=process_images, inputs=inputt, outputs=outputs, title="Multiple Object Segmentation with API Integration")
# application.launch()