Image_Processing/stream.py at master · Pri1235/Image_Processing · GitHub

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
import streamlit as st

import numpy as np
from PIL import Image
import cv2
import numpy as np
from PIL import Image
import matplotlib.pyplot as plt
import io


blur = np.array([[
    [0.0625,0.125,0.0625],
    [0.125,0.25,0.125],
    [0.0625,0.125,0.0625]
]])

edge = np.array([
    [-1,-1,-1],
    [-1,8,-1],
    [-1,-1,-1]
])
sharpen = np.array([
    [0,-1,0],
    [-1,5,-1],
    [0,-1,0]
])

v_edge = np.array([
    [-1,0,1],
    [-1,0,1],
    [-1,0,1]
])

h_edge = np.array([
    [-1,-1,-1],
    [0,0,0],
    [1,1,1]
])

bblur = np.array([
    [1/9,1/9,1/9],
    [1/9,1/9,1/9],
    [1/9,1/9,1/9]
])


def convolve(img: np.array,kernel: np.array):

    height,width,c = img.shape[0],img.shape[1],img.shape[2]

    K_height,K_width = kernel.shape[0],kernel.shape[1]

    convolved_img = np.zeros((height-K_height+1,width-K_width+1,3))

    #with padding
    for i in range(K_height//2,height-K_height//2 -1):
        for j in range(K_width//2,width-K_width//2-1):

            subset = img[i-K_height//2:i+K_height//2+1,j-K_width//2:j+K_width//2+1]

            convolved_img[i,j,0] = int((subset[:,:,0]*kernel).sum())
            convolved_img[i,j,1] = int((subset[:,:,1]*kernel).sum())
            convolved_img[i,j,2] = int((subset[:,:,2]*kernel).sum())

    convolved_img = np.clip(convolved_img,0,255)
    return convolved_img.astype(np.uint8)


def show_img(img: str,kernel: str)->np.array:
    # if kernel == 'blur':
    #     kernel = blur
    # elif kernel == 'edge':
    #     kernel = edge
    # else:
    #     kernel = sharpen
    kernel_map = {
        'blur':blur,
        'BoxBlur':bblur,
        'edge':edge,
        'sharpen':sharpen,
        'VerticalEdge':v_edge,
        'HorizontalEdge':h_edge
    }
    selected_kernel = kernel_map.get(kernel,sharpen)
    img = Image.open(img)
    img = np.asarray(img)
    image = convolve(img,selected_kernel)
    # image = Image.fromarray(image)
    # image.show()
    return image

def main():
    st.title("Play with Kernels!")
    st.divider()
    st.subheader("Choose Image")
    data = st.file_uploader("upload an Image!")
    st.subheader("Select a kernel")
    # choice1 = st.radio("Pick kernel size",["3","5","7"])
    choice = st.selectbox("PickOne",["blur","edge","sharpen","BoxBlur","VerticalEdge","HorizontalEdge"])
    st.write("You selected kernel of size: 3 and type: ",choice)
    # st.subheader("Or take a picture")
    # image = st.camera_input("take picture")

    st.subheader("Output!")
    if data is not None:
        if st.button("process") is not None:
            file_bytes = np.asarray(bytearray(data.read()),dtype=np.uint8)
        # show_img(file_bytes,choice)
            image1 = cv2.imdecode(file_bytes,1)
            cv2.imwrite('image_path.jpg', image1)

        # Use the image file path as needed
            image_path = 'image_path.jpg'
            output = show_img(image_path,choice)
            output_rgb = cv2.cvtColor(output, cv2.COLOR_BGR2RGB)
            _, encoded_image = cv2.imencode('.png', output_rgb)
            image_bytes = encoded_image.tobytes()

# Convert the image bytes to a BytesIO object
            image_io = io.BytesIO(image_bytes)

# Display the image in Streamlit
            st.image(image_io, channels="RGB")
            st.download_button("Download Image", data=image_bytes, mime="image/png")
        # st.image(output,channels="BGR")

        # st.pyplot(p)
        # p.savefig("output.png")
        # st.download_button("Download Image")
main()