streamlit_app.py

import numpy as np
from sklearn.cluster import KMeans
import matplotlib.pyplot as plt
import cv2
from collections import Counter
import streamlit as st
import time


def RGB2HEX(color):
    return "#{:02x}{:02x}{:02x}".format(int(color[0]), int(color[1]), int(color[2]))


def get_colors(image, mc):
    length = int(600 * float(image.shape[0] / image.shape[1]))
    image = cv2.resize(image, (600, length), interpolation=cv2.INTER_AREA)
    # st.image(image)
    modified_image = image.reshape(image.shape[0] * image.shape[1], 3)
    clf = KMeans(n_clusters=mc, random_state=42)
    labels = clf.fit_predict(modified_image)

    counts = Counter(labels)

    center_colors = clf.cluster_centers_
    ordered_colors = [center_colors[i] for i in counts.keys()]
    hex_colors = [RGB2HEX(ordered_colors[i]) for i in counts.keys()]

    fig1, ax1 = plt.subplots()

    def make_autopct():
        def my_autopct(pct):
            return '{p:.2f}%'.format(p=pct)
        return my_autopct

    ax1.pie(counts.values(), labels=hex_colors, colors=hex_colors, autopct=make_autopct(),
            rotatelabels=True, wedgeprops={'animated': True, "edgecolor": "black", 'linewidth': 0.05,
                                           'antialiased': True}, textprops={'size': 5, 'color': "black"})
    fig1.patch.set_facecolor(color="None")
    txt = "Identified colours"
    fig1.text(0.515, 0.06, txt, wrap=True, horizontalalignment='center', fontsize=5.5, color='grey',
              fontfamily='sans-serif', fontweight='ultralight')
    return fig1


if __name__ == "__main__":
    menu_items = {
        'About': '''
        ### Colour Identification in Images 
        ##### Made by - Rohan Karan \n 
        [![LinkedIn Follow](
        https://img.shields.io/badge/LinkedIn-0077B5?style=for-the-badge&logo=linkedin&logoColor=white)](
        https://linkedin.com/in/rohankaran001) 
        '''
    }
    st.set_page_config(page_title="Colour Identification in Images",
                       page_icon="https://www.dropbox.com/s/h2dcp308pz8w34h/rgb%20%281%29.png?raw=1",
                       menu_items=menu_items)
    st.markdown("""
        <style>
        footer {visibility: hidden;}
        </style>""", unsafe_allow_html=True)

    st.title("Colour Identification in Images")

    uploaded_file = st.file_uploader("Choose an image file", type=["jpg", "jpeg", "png"])

    if uploaded_file is not None:
        file_bytes = np.asarray(bytearray(uploaded_file.read()), dtype=np.uint8)
        opencv_image = cv2.imdecode(file_bytes, cv2.IMREAD_COLOR)
        up_image = cv2.cvtColor(opencv_image, cv2.COLOR_BGR2RGB)
        st.subheader("Uploaded image:")
        st.image(up_image, caption="Uploaded image")
        max_unique_colors = len(np.unique(up_image.reshape(-1, up_image.shape[-1]), axis=0, return_counts=True)[0])
        max_value = min(15, max_unique_colors)
        st.sidebar.subheader("Maximum colours:")
        if max_unique_colors <= 1:
            max_colors = 1
            st.sidebar.warning("Only one colour!")
        else:
            max_colors = st.sidebar.slider(f'Choose between 1-{max_value}', min_value=1, max_value=max_value,
                                           value=max_value//2 + 1)

        with st.spinner("Analyzing..."):
            load = st.image("https://www.dropbox.com/s/aejzhvqvtegnp02/Spinner-1s-800px.svg?raw=1")
            start = time.time()
            fig = get_colors(up_image, max_colors)
            end = time.time()
        load.empty()
        st.header("\n")
        st.subheader("Identified colours:")
        st.pyplot(fig)
        st.success("Done! Colours are identified using K-means clustering algorithm in {:.2f} seconds".format(end - start))
        st.balloons()