The funtion of colormap2label may not be effective , the converted label does not correspond.

[lihua7351]

test-rgb2label.pdf The result for [0,0,128] should be labeled as 4, but got the 0. Hope to get your reply, thank you! The input image from VOC2012\SegmentationClass\2007_000061.png

[lihua7351]

` import os import json #import cv2 from tqdm import tqdm import numpy as np from glob import glob import tensorflow as tf import matplotlib.pyplot as plt import skimage.io as io from tensorflow.keras.callbacks import Callback from tensorflow.keras.callbacks import TensorBoard, ModelCheckpoint, EarlyStopping, LearningRateScheduler print('TensorFlow', tf.version)

img1 = tf.io.read_file("C:/Users/eadhaw/Desktop/1221/5.png") img1 = tf.image.decode_png(img1) print(img1.shape)

VOC_COLORMAP = [[0, 0, 0], [128, 0, 0], [0, 128, 0], [128, 128, 0], [0, 0, 128], [128, 0, 128], [0, 128, 128], [128, 128, 128], [64, 0, 0], [192, 0, 0], [64, 128, 0], [192, 128, 0], [64, 0, 128], [192, 0, 128], [64, 128, 128], [192, 128, 128], [0, 64, 0], [128, 64, 0], [0, 192, 0], [128, 192, 0], [0, 64, 128]] colormap2label = np.zeros(256 ** 3, dtype=np.uint8)

for i, colormap in enumerate(VOC_COLORMAP): colormap2label[(colormap[0] * 256 + colormap[1]) * 256 + colormap[2]] = i colormap2label = tf.convert_to_tensor(colormap2label) def voc_label_indices(colormap, colormap2label): """ convert colormap (tf image) to colormap2label (uint8 tensor). """ colormap = tf.cast(colormap, dtype=tf.int32) idx = tf.add(tf.multiply(colormap[:, :, 0], 256), colormap[:, :, 1]) idx = tf.add(tf.multiply(idx, 256), colormap[:, :, 2]) idx = tf.add(idx, colormap[:, :, 2]) return tf.gather_nd(colormap2label, tf.expand_dims(idx, -1))

y = voc_label_indices(img1, colormap2label)

img1[230:240,325:330,:]

y[230:240,325:330]`

[SwordFaith]

Which chapter is this question about ?

[lihua7351]

chapter09_computer-vision/9.9_semantic-segmentation-and-dataset.ipynb/def voc_label_indices(colormap, colormap2label):

[lihua7351]

这个问题是关于哪一章的？ Chapter09_computer-vision / 9.9_semantic-segmentation-and-dataset.ipynb