File size: 3,643 Bytes
cec5e36
 
 
 
6d8dbab
 
 
0ba2d7d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
a34a00e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cec5e36
 
 
 
0cb35c3
 
 
 
3d7753d
0cb35c3
 
 
 
 
 
 
 
0b3216f
0cb35c3
eb64ea1
 
 
 
 
 
 
 
 
 
 
 
 
cec5e36
 
eb64ea1
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cec5e36
 
 
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
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
from huggingface_hub import from_pretrained_fastai
import gradio as gr
from fastai.vision.all import *

def get_y_fn (x):
    return Path(str(x).replace("Images","Labels").replace("color","gt").replace(".jpg",".png"))

from albumentations import (
    Compose,
    OneOf,
    ElasticTransform,
    GridDistortion,
    OpticalDistortion,
    HorizontalFlip,
    Rotate,
    Transpose,
    CLAHE,
    ShiftScaleRotate,
    RandomBrightnessContrast,
    GaussNoise
)

class SegmentationAlbumentationsTransform(ItemTransform):
    split_idx = 0

    def __init__(self, aug):
        self.aug = aug

    def encodes(self, x):
        img,mask = x
        aug = self.aug(image=np.array(img), mask=np.array(mask))
        return PILImage.create(aug["image"]), PILMask.create(aug["mask"])

class TargetMaskConvertTransform(ItemTransform):
    def __init__(self):
        pass
    def encodes(self, x):
        img,mask = x

        #Convert to array
        mask = np.array(mask)

        # Uvas
        mask[mask==255]=1

        # Hojas
        mask[mask==150]=2

        # Poste
        mask[mask==76]=3
        mask[mask==74]=3

        # Madera
        mask[mask==29]=4
        mask[mask==25]=4

        # Back to PILMask
        mask = PILMask.create(mask)
        return img, mask

repo_id = "ancebuc/grapes-segmentation"

learner = from_pretrained_fastai(repo_id)

aux = learner.model
aux = aux.cpu()

import torchvision.transforms as transforms
img = PILImage.create('color_158.jpg')
transformer=transforms.Compose([transforms.Resize((480,640)),
                                transforms.ToTensor(),
                                transforms.Normalize(
                                                    [0.485, 0.456, 0.406],
                                                    [0.229, 0.224, 0.225])])
img=transformer(img).unsqueeze(0)
img=img.cpu()

model=torch.jit.trace(aux, (img))

device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
model = model.cpu()
model.eval()

import torchvision.transforms as transforms
def transform_image(image):
    my_transforms = transforms.Compose([transforms.ToTensor(),
                                        transforms.Normalize(
                                            [0.485, 0.456, 0.406],
                                            [0.229, 0.224, 0.225])])
    image_aux = image
    return my_transforms(image_aux).unsqueeze(0).to(device)

# Definimos una función que se encarga de llevar a cabo las predicciones
def predict(img):
    img = PILImage.create(img)

    image = transforms.Resize((480,640))(img)
    tensor = transform_image(image=image)

    with torch.no_grad():
        outputs = model(tensor)

    outputs = torch.argmax(outputs,1)

    mask = np.array(outputs.cpu())
    mask = np.reshape(mask,(480,640))

    # Añadimos una dimesionalidad para colocar color
    mask = np.expand_dims(mask, axis=2)

    # Y añadimos los tres canales
    mask = np.repeat(mask, 3, axis=2)

    # Creamos las máscaras
    uvas = np.all(mask == [1, 1, 1], axis=2)
    hojas = np.all(mask == [2, 2, 2], axis=2)
    poste = np.all(mask == [3, 3, 3], axis=2)
    madera = np.all(mask == [4, 4, 4], axis=2)
    
    # Uvas
    mask[uvas] = [255, 255, 255]
    
    # Hojas
    mask[hojas] = [0, 255, 0]
    
    # Poste
    mask[poste] = [0, 0, 255]
    
    # Madera
    mask[madera] = [255, 0, 0]

    return Image.fromarray(mask.astype('uint8'))
    
# Creamos la interfaz y la lanzamos. 
gr.Interface(fn=predict, inputs=gr.inputs.Image(shape=(128, 128)), outputs=gr.outputs.inputs.Image(shape=(128, 128)),examples=['color_158.jpg','color_157.jpg']).launch(share=False)