用于 PNG、GIF 等的 Tensorflow Label_Image

Question

我正在使用 Tensorflow 进行图像分类。我使用 image_retraining/retrain.py 使用新类别重新训练初始库，并使用它使用 https://github.com/llSourcell/tensorflow_image_classifier/blob/master/src/label_image.py 中的 label_image.py 对图像进行分类，如下所示：

import tensorflow as tf
import sys

# change this as you see fit
image_path = sys.argv[1]

# Read in the image_data
image_data = tf.gfile.FastGFile(image_path, 'rb').read()

# Loads label file, strips off carriage return
label_lines = [line.rstrip() for line
    in tf.gfile.GFile("/root/tf_files/output_labels.txt")]

# Unpersists graph from file
with tf.gfile.FastGFile("/root/tf_files/output_graph.pb", 'rb') as f:
    graph_def = tf.GraphDef()
    graph_def.ParseFromString(f.read())
    _ = tf.import_graph_def(graph_def, name='')

with tf.Session() as sess:
# Feed the image_data as input to the graph and get first prediction
    softmax_tensor = sess.graph.get_tensor_by_name('final_result:0')

    #predictions = sess.run(softmax_tensor,{'DecodeJpeg/contents:0': image_data})
        predictions = sess.run(softmax_tensor,{'DecodePng/contents:0': image_data})

# Sort to show labels of first prediction in order of confidence
    top_k = predictions[0].argsort()[-len(predictions[0]):][::-1]

    for node_id in top_k:
        human_string = label_lines[node_id]
        score = predictions[0][node_id]
        print('%s (score = %.5f)' % (human_string, score))

我注意到两个问题。当我用新类别重新训练时，它只训练 JPG 图像。我是机器学习的菜鸟，所以不确定这是否是一个限制，或者是否可以训练其他扩展图像，如 PNG、GIF？

另一个是在对图像进行分类时，输入再次仅适用于 JPG。我试图在上面的 label_image.py 中将 DecodeJpeg 更改为 DecodePng 但无法工作。我尝试的另一种方法是将其他格式转换为 JPG，然后再将它们传递给分类，例如：

im = Image.open('/root/Desktop/200_s.gif').convert('RGB')
im.save('/root/Desktop/test.jpg', "JPEG")
image_path1 = '/root/Desktop/test.jpg'

还有其他方法吗？ Tensorflow 是否具有处理 JPG 以外的其他图像格式的功能？

我按照@mrry

的建议，通过输入与 JPEG 相比的解析图像来尝试以下操作

import tensorflow as tf
import sys
import numpy as np
from PIL import Image

# change this as you see fit
image_path = sys.argv[1]

# Read in the image_data
image_data = tf.gfile.FastGFile(image_path, 'rb').read()

image = Image.open(image_path)
image_array = np.array(image)[:,:,0:3]  # Select RGB channels only.

# Loads label file, strips off carriage return
label_lines = [line.rstrip() for line
    in tf.gfile.GFile("/root/tf_files/output_labels.txt")]

# Unpersists graph from file
with tf.gfile.FastGFile("/root/tf_files/output_graph.pb", 'rb') as f:
    graph_def = tf.GraphDef()
    graph_def.ParseFromString(f.read())
    _ = tf.import_graph_def(graph_def, name='')

with tf.Session() as sess:
# Feed the image_data as input to the graph and get first prediction
    softmax_tensor = sess.graph.get_tensor_by_name('final_result:0')

        predictions = sess.run(softmax_tensor,{'DecodeJpeg:0': image_array})

# Sort to show labels of first prediction in order of confidence
    top_k = predictions[0].argsort()[-len(predictions[0]):][::-1]

    for node_id in top_k:
        human_string = label_lines[node_id]
        score = predictions[0][node_id]
        print('%s (score = %.5f)' % (human_string, score))

它适用于 JPEG 图片，但当我使用 PNG 或 GIF 时它抛出

        Traceback (most recent call last):   
    File "label_image.py", line 17, in <module>
        image_array = np.array(image)[:,:,0:3]  # Select RGB channels only. 
IndexError: too many indices for array

Answer 1

您应该看看 tf.image 包。它具有解码/编码 JPEG、GIF 和 PNG 的良好功能。

Answer 2

该模型只能训练（和评估）JPEG 图像，因为您保存在 /root/tf_files/output_graph.pb 中的 GraphDef 仅包含一个 tf.image.decode_jpeg() 操作，并使用输出该操作用于进行预测。使用其他图像格式至少有几个选项：

输入经过解析的图像而不是 JPEG 数据。 在当前程序中，您输入 JPEG 编码图像作为张量的字符串值 "DecodeJpeg/contents:0"。相反，您可以为张量 "DecodeJpeg:0"（表示 tf.image.decode_jpeg() 操作的输出）输入解码图像数据的 3-D 数组，然后您可以使用 NumPy、PIL 或其他一些 Python 库来创建此数组。

重新映射 tf.import_graph_def() 中的图像输入。 tf.import_graph_def() 函数使您能够通过重新映射单个张量将两个不同的图形连接在一起值。例如，您可以执行类似以下操作以向现有图形添加新的图像处理操作：

image_string_input = tf.placeholder(tf.string)
image_decoded = tf.image.decode_png(image_string_input)

# Unpersists graph from file
with tf.gfile.FastGFile("/root/tf_files/output_graph.pb", 'rb') as f:
    graph_def = tf.GraphDef()
    graph_def.ParseFromString(f.read())
    softmax_tensor, = tf.import_graph_def(
        graph_def,
        input_map={"DecodeJpeg:0": image_decoded},
        return_operations=["final_result:0"])

with tf.Session() as sess:
    # Feed the image_data as input to the graph and get first prediction
    predictions = sess.run(softmax_tensor, {image_string_input: image_data})
    # ...

Answer 3

按照@mrry 的建议输入解析后的图像，将图像数据转换为数组并转换为 RGB，如下面的代码所述。现在我可以输入 JPG、PNG 和 GIF。

import tensorflow as tf
import sys
import numpy as np
from PIL import Image

# change this as you see fit
image_path = sys.argv[1]

# Read in the image_data
image_data = tf.gfile.FastGFile(image_path, 'rb').read()

image = Image.open(image_path)
image_array = image.convert('RGB')

# Loads label file, strips off carriage return
label_lines = [line.rstrip() for line
    in tf.gfile.GFile("/root/tf_files/output_labels.txt")]

# Unpersists graph from file
with tf.gfile.FastGFile("/root/tf_files/output_graph.pb", 'rb') as f:
    graph_def = tf.GraphDef()
    graph_def.ParseFromString(f.read())
    _ = tf.import_graph_def(graph_def, name='')

with tf.Session() as sess:
# Feed the image_data as input to the graph and get first prediction
    softmax_tensor = sess.graph.get_tensor_by_name('final_result:0')

        predictions = sess.run(softmax_tensor,{'DecodeJpeg:0': image_array})

# Sort to show labels of first prediction in order of confidence
    top_k = predictions[0].argsort()[-len(predictions[0]):][::-1]

    for node_id in top_k:
        human_string = label_lines[node_id]
        score = predictions[0][node_id]
        print('%s (score = %.5f)' % (human_string, score))

用于 PNG、GIF 等的 Tensorflow Label_Image

Tensorflow Label_Image for PNG, GIF etc

python

jpeg

classification

image-processing

tensorflow