[RoboND_recognition] #recognition #robotics

001_recongnition_overview.txt

purpose:

to give each cluster in segmentation output a label

s.t. the grounding geometric specification of semantic command can be extracted

002_color_space.py

#hsv: hue saturation and vlue

#value stands for brightness

#color histogram: 
#the bag of words representation of color
#divide each color channel into bins and quantize each pixel
#sum over all pixels to get BOW representation

import matplotlib.image as mpimg
import matplotlib.pyplot as plt
import numpy as np

image = mpimg.imread('can.jpeg')
plt.imshow(image)


# Take histograms in R, G, and B
r_hist = np.histogram(image[:,:,0], bins=32, range=(0, 256))
g_hist = np.histogram(image[:,:,1], bins=32, range=(0, 256))
b_hist = np.histogram(image[:,:,2], bins=32, range=(0, 256))

bin_edges = r_hist[1]
bin_centers = (bin_edges[1:]  + bin_edges[0:len(bin_edges)-1])/2

# Plot a figure with all three bar charts
fig = plt.figure(figsize=(12,3))
plt.subplot(131)
plt.bar(bin_centers, r_hist[0])
plt.xlim(0, 256)
plt.title('R Histogram')
plt.subplot(132)
plt.bar(bin_centers, g_hist[0])
plt.xlim(0, 256)
plt.title('G Histogram')
plt.subplot(133)
plt.bar(bin_centers, b_hist[0])
plt.xlim(0, 256)
plt.title('B Histogram')
plt.show()

hist_features = np.concatenate((r_hist[0], g_hist[0], b_hist[0])).astype(np.float64)

003_hsv_histogram.py

import numpy as np
import cv2
import matplotlib.pyplot as plt
import matplotlib.image as mpimg

def color_hist(img, nbins=32, bins_range=(0, 256)):
    # Convert from RGB to HSV using cv2.cvtColor()
    # Compute the histogram of the HSV channels separately
    # Concatenate the histograms into a single feature vector
    # Normalize the result
    # Return the feature vector
    image = cv2.cvtColor(img, cv2.COLOR_RGB2HSV)
    r_hist = np.histogram(image[:,:,0], bins=nbins, range=bins_range)
    g_hist = np.histogram(image[:,:,1], bins=nbins, range=bins_range)
    b_hist = np.histogram(image[:,:,2], bins=nbins, range=bins_range)
    hist_features = np.concatenate((r_hist[0], g_hist[0], b_hist[0])).astype(np.float64)
    norm_features = hist_features / np.sum(hist_features)
    return norm_features