captura2.py

# Importando las librerías Gradio, requests, PIL e io
import gradio as gr
import requests
from PIL import Image
from io import BytesIO
import argparse
import os
import cv2
import requests
import numpy as np
from pathlib import Path
import warnings

import torch

from groundingdino.models import build_model
from groundingdino.util.slconfig import SLConfig
from groundingdino.util.utils import clean_state_dict
from groundingdino.util.inference import annotate, load_image, predict
import groundingdino.datasets.transforms as T

from huggingface_hub import hf_hub_download

# Use this command for evaluate the GLIP-T model
config_file = "groundingdino/config/GroundingDINO_SwinT_OGC.py"
ckpt_repo_id = "ShilongLiu/GroundingDINO"
ckpt_filenmae = "groundingdino_swint_ogc.pth"

def load_model_hf(model_config_path, repo_id, filename, device='cpu'):
    args = SLConfig.fromfile(model_config_path) 
    model = build_model(args)
    args.device = device

    cache_file = hf_hub_download(repo_id=repo_id, filename=filename)
    checkpoint = torch.load(cache_file, map_location='cpu')
    log = model.load_state_dict(clean_state_dict(checkpoint['model']), strict=False)
    print("Model loaded from {} \n => {}".format(cache_file, log))
    _ = model.eval()
    return model    

def image_transform_grounding(init_image):
    transform = T.Compose([
        T.RandomResize([800], max_size=1333),
        T.ToTensor(),
        T.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
    ])
    image, _ = transform(init_image, None) # 3, h, w
    return init_image, image

def image_transform_grounding_for_vis(init_image):
    transform = T.Compose([
        T.RandomResize([800], max_size=1333),
    ])
    image, _ = transform(init_image, None) # 3, h, w
    return image

model = load_model_hf(config_file, ckpt_repo_id, ckpt_filenmae)

def run_grounding(input_image, grounding_caption, box_threshold, text_threshold):
    init_image = input_image.convert("RGB")
    original_size = init_image.size

    _, image_tensor = image_transform_grounding(init_image)
    image_pil: Image = image_transform_grounding_for_vis(init_image)

    # run grounding
    boxes, logits, phrases = predict(model, image_tensor, grounding_caption, box_threshold, text_threshold, device='cpu')
    annotated_frame = annotate(image_source=np.asarray(image_pil), boxes=boxes, logits=logits, phrases=phrases)
    image_with_box = Image.fromarray(cv2.cvtColor(annotated_frame, cv2.COLOR_BGR2RGB))

    return image_with_box


# Definiendo la función captura_pagina
def captura_pagina(url):
    # Asignando la clave de la API y la URL
    api_key = 'b77e9ec7b82e4447b93c73cf1af4a93f'
    api_url = f'https://api.apiflash.com/v1/urltoimage?access_key={api_key}&url={url}'

    # Haciendo una solicitud GET a la API
    respuesta = requests.get(api_url, stream=True)

    # Si la solicitud es exitosa, se procesa la imagen
    if respuesta.status_code == 200:
        image_data = b''
        for chunk in respuesta.iter_content(8192):
            image_data += chunk
        image = Image.open(BytesIO(image_data))

        # Set the fixed box_threshold and text_threshold
        box_threshold = 0.38
        text_threshold = 0.25
        grounding_caption = "Find the webform in the picture of a web."

        # Run the Grounding DINO model on the image
        image_with_bb = run_grounding(image, grounding_caption, box_threshold, text_threshold)

        return "¡Página web capturada con éxito!", image, image_with_bb
    else:
        # Si la solicitud no es exitosa, se retorna un mensaje de error
        return f'Error: {respuesta.status_code}', None, None

# Definiendo la función captura_pagina_app
def captura_pagina_app():
    # Creando un objeto de la clase Row de Gradio
    with gr.Row():
        with gr.Column():
            # Agregando un cuadro de texto para ingresar la URL
            textbox_url = gr.Textbox(label='URL')

            # Agregando un botón para capturar la página web
            btn_predecir = gr.Button(value='Predecir')
        with gr.Column():
            # Agregando un cuadro de texto para mostrar el estado
            output_mensaje = gr.Textbox(label='Estado')

            # Agregando dos imágenes para mostrar la captura de la página web
            output_img1 = gr.Image()
            output_img2 = gr.Image()

    # Asociando la función captura_pagina con el botón
    btn_predecir.click(
        fn=captura_pagina,
        inputs=textbox_url,
        outputs=[output_mensaje, output_img1, output_img2]
    )