import numpy as np
import matplotlib.pyplot as plt
def plot_image_histogram_rgb(image, nbins=64):
# Check if the image is grayscale or RGB
if len(image.shape) == 2: # Grayscale image
# Grayscale histogram
plt.figure(figsize=(12, 3))
plt.hist(image.ravel(), bins=nbins, color='gray', edgecolor='black', linewidth=0.5, alpha=0.6, label='Grayscale Histogram')
plt.xlabel('Pixel Value')
plt.ylabel('Frequency')
plt.title('Grayscale Histogram')
plt.xlim([0, 255])
plt.legend()
plt.tight_layout()
plt.show()
elif len(image.shape) == 3 and image.shape[2] == 3: # RGB image
# Separate the channels
red_channel = image[:, :, 0]
green_channel = image[:, :, 1]
blue_channel = image[:, :, 2]
# Plot histograms for each channel
plt.figure(figsize=(12, 3))
# Red channel histogram
plt.hist(red_channel.ravel(), bins=nbins, range=(0, 255), color='red', edgecolor='black', linewidth=0.5, alpha=0.6, label='Red Channel Histogram')
# Green channel histogram
plt.hist(green_channel.ravel(), bins=nbins, range=(0, 255), color='green', edgecolor='black', linewidth=0.5, alpha=0.6, label='Green Channel Histogram')
# Blue channel histogram
plt.hist(blue_channel.ravel(), bins=nbins, range=(0, 255), color='blue', edgecolor='black', linewidth=0.5, alpha=0.6, label='Blue Channel Histogram')
plt.xlabel('Pixel Value')
plt.ylabel('Frequency')
plt.title('Color Channel Histograms')
plt.xlim([0, 255])
plt.legend()
plt.tight_layout()
plt.show()
else:
print("Unsupported image format")
def plot_image_histogram_hue(image, nbins=64):
# Convert the image to HSV color space
hsv_image = cv2.cvtColor(image, cv2.COLOR_RGB2HSV)
# Separate the channels
hue_channel = hsv_image[:, :, 0]
# Plot histograms for each channel
plt.figure(figsize=(12, 3))
# Hue channel histogram with corresponding color
hue_hist, bins = np.histogram(hue_channel.ravel(), bins=nbins, range=[0, 180])
bin_centers = (bins[:-1] + bins[1:]) / 2
# Normalize the bin centers to map them to [0, 1] for colors
norm_bin_centers = bin_centers / 180
# Create a color map based on the HSV values
colors = plt.cm.hsv(norm_bin_centers)
for i in range(nbins):
plt.bar(bin_centers[i]*2, hue_hist[i], color=colors[i], width=((bins[1] - bins[0])* 2), edgecolor='black', alpha=0.6)
plt.title('Hue Channel Histogram with Corresponding Colors')
plt.xlabel('Hue Value (Degree)')
plt.ylabel('Frequency')
plt.xlim([0, 360])
plt.tight_layout()
plt.show()
def plot_image_histogram_saturation(image, nbins=64):
# Convert the image to HSV color space
hsv_image = cv2.cvtColor(image, cv2.COLOR_RGB2HSV)
# Extract the saturation channel
saturation_channel = hsv_image[:, :, 1]
plt.figure(figsize=(12, 3))
# Saturation channel histogram (black to white gradient)
saturation_hist, bins = np.histogram(saturation_channel.ravel(), bins=nbins, range=[0, 255])
bin_centers = (bins[:-1] + bins[1:]) / 2
# Create a color map from blue to white for value
colors_saturation = plt.cm.Blues(bin_centers / 255)
for i in range(nbins):
plt.bar(bin_centers[i], saturation_hist[i], color=colors_saturation[i], width=(bins[1] - bins[0]), edgecolor='black', alpha=0.6)
plt.title('Saturation Channel Histogram (Blue to White Gradient)')
plt.xlabel('Saturation')
plt.ylabel('Frequency')
plt.xlim([0, 255])
plt.tight_layout()
plt.show()
def plot_image_histogram_value(image, nbins=64):
# Convert the image to HSV color space
hsv_image = cv2.cvtColor(image, cv2.COLOR_RGB2HSV)
# Extract the value channel
value_channel = hsv_image[:, :, 2]
plt.figure(figsize=(12, 3))
# Value channel histogram (black to white gradient)
value_hist, bins = np.histogram(value_channel.ravel(), bins=nbins, range=[0, 255])
bin_centers = (bins[:-1] + bins[1:]) / 2
# Create a color map from black to white for value
colors_value = plt.cm.Greys(bin_centers / 255)
for i in range(nbins):
plt.bar(bin_centers[i], value_hist[i], color=colors_value[i], width=(bins[1] - bins[0]), edgecolor='black', alpha=0.6)
plt.title('Value Channel Histogram (Black to White Gradient)')
plt.xlabel('Value')
plt.ylabel('Frequency')
plt.xlim([0, 255])
plt.tight_layout()
plt.show()
def uint82float(image_uint8, is_hsv=False):
"""
Convert an image from uint8 [0, 255] to float [0.0, 1.0]
Parameters:
image_uint8 (numpy array): Input image in uint8 format
is_hsv (Boolean): If the image is in HSV format, scale H differently (0-180 for H, 0-255 for S and V)
Returns:
numpy array: Image converted to float format
"""
if is_hsv:
# H channel is scaled to [0, 1] from its range of [0, 180]
h_float = image_uint8[:, :, 0].astype(np.float32) / 180.0
# S and V channels are scaled to [0, 1] from their range of [0, 255]
s_float = image_uint8[:, :, 1].astype(np.float32) / 255.0
v_float = image_uint8[:, :, 2].astype(np.float32) / 255.0
# Stack the H, S, and V channels back into a single array
return np.stack((h_float, s_float, v_float), axis=2)
else:
# For RGB images, scale all channels uniformly
return image_uint8.astype(np.float32) / 255.0
def float2uint8(image_float, is_hsv=False):
"""
Convert an image from float [0.0, 1.0] to uint8 [0, 255] (for RGB) or [0, 180] (for H channel in HSV)
Parameters:
image_float (numpy array): Input image in float format
is_hsv (Boolean): If the image is in HSV format, scale H channel differently
Returns:
numpy array: Image converted to uint8 format
"""
if is_hsv:
# H channel is scaled back to its original range of [0, 180]
h_uint8 = np.clip(image_float[:, :, 0] * 180, 0, 180).astype(np.uint8)
# S and V channels are scaled back to [0, 255]
s_uint8 = np.clip(image_float[:, :, 1] * 255, 0, 255).astype(np.uint8)
v_uint8 = np.clip(image_float[:, :, 2] * 255, 0, 255).astype(np.uint8)
# Stack the H, S, and V channels back into a single array
return np.stack((h_uint8, s_uint8, v_uint8), axis=2)
else:
# For RGB images, scale all channels uniformly
return np.clip(image_float * 255, 0, 255).astype(np.uint8)
def uint82float(image_uint8, is_hsv=False):
"""
Convert an image from uint8 [0, 255] to float [0.0, 1.0]
Parameters:
image_uint8 (numpy array): Input image in uint8 format
is_hsv (Boolean): If the image is in HSV format, scale H differently (0-180 for H, 0-255 for S and V)
Returns:
numpy array: Image converted to float format
"""
if is_hsv:
# H channel is scaled to [0, 1] from its range of [0, 180]
h_float = image_uint8[:, :, 0].astype(np.float32) / 180.0
# S and V channels are scaled to [0, 1] from their range of [0, 255]
s_float = image_uint8[:, :, 1].astype(np.float32) / 255.0
v_float = image_uint8[:, :, 2].astype(np.float32) / 255.0
# Stack the H, S, and V channels back into a single array
return np.stack((h_float, s_float, v_float), axis=2)
else:
# For RGB images, scale all channels uniformly
return image_uint8.astype(np.float32) / 255.0
def float2uint8(image_float, is_hsv=False):
"""
Convert an image from float [0.0, 1.0] to uint8 [0, 255] (for RGB) or [0, 180] (for H channel in HSV)
Parameters:
image_float (numpy array): Input image in float format
is_hsv (Boolean): If the image is in HSV format, scale H channel differently
Returns:
numpy array: Image converted to uint8 format
"""
if is_hsv:
# H channel is scaled back to its original range of [0, 180]
h_uint8 = np.clip(image_float[:, :, 0] * 180, 0, 180).astype(np.uint8)
# S and V channels are scaled back to [0, 255]
s_uint8 = np.clip(image_float[:, :, 1] * 255, 0, 255).astype(np.uint8)
v_uint8 = np.clip(image_float[:, :, 2] * 255, 0, 255).astype(np.uint8)
# Stack the H, S, and V channels back into a single array
return np.stack((h_uint8, s_uint8, v_uint8), axis=2)
else:
# For RGB images, scale all channels uniformly
return np.clip(image_float * 255, 0, 255).astype(np.uint8)