File size: 2,632 Bytes
03a7dca 42e8595 03a7dca 2a13288 03a7dca 2a13288 03a7dca 2a13288 03a7dca 2a13288 03a7dca 2a13288 03a7dca 2a13288 03a7dca 2a13288 03a7dca 2a13288 03a7dca 2a13288 03a7dca |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 |
import gradio as gr
import torch
from PIL import Image
from gtts import gTTS
import numpy as np
import cv2
from sklearn.feature_extraction.image import greycomatrix, greycoprops
from transformers import BlipProcessor, BlipForConditionalGeneration, MarianMTModel, MarianTokenizer
# Carregar o modelo YOLOv5
model = torch.hub.load('ultralytics/yolov5', 'yolov5s')
# Função para análise de textura usando GLCM
def analyze_texture(image):
gray_image = cv2.cvtColor(np.array(image), cv2.COLOR_BGR2GRAY)
glcm = greycomatrix(gray_image, distances=[5], angles=[0], levels=256, symmetric=True, normed=True)
contrast = greycoprops(glcm, 'contrast')[0, 0]
return contrast
# Função para descrever imagem usando BLIP
def describe_image(image):
processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-base")
model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-base")
inputs = processor(image, return_tensors="pt")
out = model.generate(**inputs)
description = processor.decode(out[0], skip_special_tokens=True)
return description
# Função para traduzir descrição para português
def translate_description(description):
model_name = 'Helsinki-NLP/opus-mt-en-pt'
tokenizer = MarianTokenizer.from_pretrained(model_name)
model = MarianMTModel.from_pretrained(model_name)
translated = model.generate(**tokenizer(description, return_tensors="pt", padding=True))
translated_text = tokenizer.decode(translated[0], skip_special_tokens=True)
return translated_text
# Função principal para processar imagem e gerar saída de voz
def process_image(image):
# Detecção de objetos
results = model(image)
detected_image = results.render()[0]
# Análise de cor (média RGB)
mean_rgb = np.mean(np.array(image), axis=(0, 1))
# Análise de textura
texture_contrast = analyze_texture(image)
# Descrição da imagem
description = describe_image(image)
translated_description = translate_description(description)
# Texto para voz
tts = gTTS(text=translated_description, lang='pt')
tts.save("output.mp3")
# Retornar imagem com detecções, descrição e áudio
return Image.fromarray(detected_image), translated_description, "output.mp3"
# Carregar imagem de exemplo
example_image = Image.open("/mnt/data/example1.JPG")
# Interface Gradio
iface = gr.Interface(
fn=process_image,
inputs=gr.inputs.Image(type="pil"),
outputs=[gr.outputs.Image(type="pil"), gr.outputs.Textbox(), gr.outputs.Audio(type="file")],
examples=[example_image]
)
iface.launch()
|