CODEWITHSUNDEEP
tensorflow
Necromancer
Necromancer

Reputation: 919

Convert RGBA-image to black and white image in Tensorflow

tf.image.decode_png() can output grayscale, RGB and RGBA image.

But I'd like to convert RGBA to pure black and white in Tensorflow (without using pillow).

Please give me some advice.

Upvotes: 2

Views: 5298

Answers (1)

Yaroslav Bulatov
Yaroslav Bulatov

Reputation: 57973

Use tf.select to do the thresholding

pil_image = PilImage.open('/temp/panda.png')
show_pil_image(pil_image)

enter image description here

pil_buf = open('/temp/panda.png').read()
contents = tf.placeholder(dtype=tf.string)
decode_op = tf.image.decode_png(contents, channels=1)
gray_image = tf.squeeze(decode_op) # shape (127,127,1) -> shape (127,127)
sess = create_session()
[decoded] = sess.run([gray_image], feed_dict={contents: pil_buf})
show_pil_image(PilImage.fromarray(decoded))

enter image description here

contents = tf.placeholder(dtype=tf.string)
decode_op = tf.image.decode_png(contents, channels=1)
gray_image = tf.squeeze(decode_op) # shape (127,127,1) -> shape (127,127)
select_op = tf.select(gray_image>127, 255*tf.ones_like(gray_image), tf.zeros_like(gray_image))
sess = create_session()
[decoded] = sess.run([select_op], feed_dict={contents: pil_buf})
show_pil_image(PilImage.fromarray(decoded))

enter image description here

Upvotes: 6

Related Questions