app.py

import warnings
from io import BytesIO

import matplotlib as mpl
import matplotlib.patches as patches
import matplotlib.pyplot as plt
import numpy as np
from mpl_toolkits.axes_grid1 import make_axes_locatable

import streamlit as st
from img2cmap import ImageConverter

warnings.filterwarnings("ignore")
st.set_option("deprecation.showfileUploaderEncoding", False)

st.set_page_config(
    page_title="img2cmap web",
    layout="wide",
)

st.title("Convert images to a colormap")
st.markdown(
    """
    This app converts images to colormaps using the Python
    library [img2cmap](https://github.com/arvkevi/img2cmap).
    Try your own image on the left. **Scroll down to generate an optimal colormap.**
    """
)

st.sidebar.markdown("### Image settings")
file_or_url = st.sidebar.radio("Upload an image file or paste an image URL", ("file", "url"))

if file_or_url == "file":
    user_image = st.sidebar.file_uploader("Upload an image file")
    if user_image is not None:
        user_image = BytesIO(user_image.getvalue())
elif file_or_url == "url":
    user_image = st.sidebar.text_input("Paste an image URL", "https://static1.bigstockphoto.com/3/2/3/large1500/323952496.jpg")
else:
    st.warning("Please select an option")

# default image to use
if user_image is None:
    user_image = "https://raw.githubusercontent.com/arvkevi/img2cmap/main/tests/images/south_beach_sunset.jpg"

# user settings
st.sidebar.markdown("### User settings")
n_colors = st.sidebar.number_input(
    "Number of colors", min_value=2, max_value=20, value=5, help="The number of colors to return in the colormap"
)
n_colors = int(n_colors)
remove_transparent = st.sidebar.checkbox(
    "Remove transparency", False, help="If checked, remove transparent pixels from the image before clustering."
)
random_state = st.sidebar.number_input("Random state", value=42, help="Random state for reproducibility")
random_state = int(random_state)


@st.cache(allow_output_mutation=True)
def get_image_converter(user_image, remove_transparent):
    converter = ImageConverter(user_image, remove_transparent=remove_transparent)
    return converter


converter = get_image_converter(user_image, remove_transparent)

with st.spinner("Generating colormap..."):
    cmap = converter.generate_cmap(n_colors=n_colors, palette_name="", random_state=random_state)

# plot the image and colorbar
fig1, ax1 = plt.subplots(figsize=(8, 8))

ax1.axis("off")
img = converter.image
im = ax1.imshow(img, cmap=cmap)

divider = make_axes_locatable(ax1)
cax = divider.append_axes("right", size="10%", pad=0.05)

cb = fig1.colorbar(im, cax=cax, orientation="vertical", label=cmap.name)
cb.set_ticks([])
st.pyplot(fig1)

colors1 = [mpl.colors.rgb2hex(c) for c in cmap.colors]
st.text("Hex Codes (click to copy on far right)")
st.code(colors1)


st.header("Detect optimal number of colors")
optimize = st.button("Optimize")
if optimize:
    with st.spinner("Optimizing... (this takes about a minute)"):
        cmaps, best_n_colors, ssd = converter.generate_optimal_cmap(max_colors=20, palette_name="", random_state=random_state)

    figopt, ax = plt.subplots(figsize=(7, 5))

    ymax = 21
    xmax = 20
    ax.set_ylim(2, ymax)
    ax.set_xlim(0, 20)

    # i will be y axis
    for y, cmap_ in cmaps.items():
        colors = sorted([mpl.colors.rgb2hex(c) for c in cmap_.colors])
        intervals, width = np.linspace(0, xmax, len(colors) + 1, retstep=True)
        # j will be x axis
        for j, color in enumerate(colors):
            rect = patches.Rectangle((intervals[j], y), width, 1, facecolor=color)
            ax.add_patch(rect)

    ax.set_yticks(np.arange(2, ymax) + 0.5)
    ax.set_yticklabels(np.arange(2, ymax))
    ax.set_ylabel("Number of colors")
    ax.set_xticks([])

    # best
    rect = patches.Rectangle((0, best_n_colors), ymax, 1, linewidth=1, facecolor="none", edgecolor="black", linestyle="--")
    ax.add_patch(rect)

    # minus 2, one for starting at 2 and one for 0-indexing
    ax.get_yticklabels()[best_n_colors - 2].set_color("red")
    st.pyplot(figopt)
    st.metric("Optimal number of colors", best_n_colors)
    st.text("Hex Codes of optimal colormap (click to copy on far right)")
    st.code(sorted([mpl.colors.rgb2hex(c) for c in cmaps[best_n_colors].colors]))

    st.text("Sum of squared distances by number of colors:")
    st.write(ssd)