代码之家 › 专栏 › 技术社区 › Ericccccc

当我使用tf时。nn。read_file()在vgg中测试图像,这是错误的。但当我用PIL打开它时,它是正确的。真奇怪

python-imaging-library tensorflow python

Ericccccc · 技术社区 · 8 年前

我编写了一个vgg模型,并对其进行了训练。现在,我要测试一个新图像。当我使用code1时,它运行良好,但在code2中是错误的。 vgg()是我定义的模型。ckpt文件已保存在“D:\Demo\ckpt”中。

代码1:它将预测打印为[1.77901700e-01 8.22093844e-01 4.42284863e-06]]

def evaluate_one_image(path):
    with tf.Graph().as_default():
        image_plt = Image.open(path)
        image = image_plt.resize([224, 224])
        image_array = np.array(image)
        image = np.reshape(image_array, (1,224,224,3))
        x = tf.placeholder(tf.float32, shape=[1, 224, 224, 3])
        logit = vgg(x)
        logit = tf.nn.softmax(logit)
        logs_train_dir = 'D:\\Demo\\ckpt'
        saver = tf.train.Saver(tf.global_variables())
        with tf.Session() as sess:
            ckpt = tf.train.get_checkpoint_state(logs_train_dir)
            saver.restore(sess, ckpt.model_checkpoint_path)               
            prediction = sess.run(logit, feed_dict={x: image})
            print(prediction)

------------------------------------------------分界线--------------------------------------

代码2:我在tensorflow中使用了一些函数。它将预测打印为[0.33333334 0.33333334 0.3333333 4]]。

def test_one_image(path):
    with tf.Graph().as_default():
        image_plt = Image.open(path)
        image_tensor = tf.image.decode_jpeg(tf.read_file(path), channels=3)
        image_tensor = tf.image.resize_image_with_crop_or_pad(image_tensor, 224, 224)
        # image_tensor = tf.image.per_image_standardization(image_tensor)
        image_tensor = tf.reshape(image_tensor, [1, 224, 224, 3])
        x = tf.placeholder(tf.float32, shape=[1, 224, 224, 3])
        logit = vgg(x)
        logit = tf.nn.softmax(logit)
        logs_train_dir = 'D:\\Demo\\ckpt'
        saver = tf.train.Saver(tf.global_variables())
        with tf.Session() as sess:
            ckpt = tf.train.get_checkpoint_state(logs_train_dir)
            saver.restore(sess, ckpt.model_checkpoint_path)              
            prediction = sess.run(logit, feed_dict={x: image_tensor.eval()})
            print(prediction)

我认为这两个代码的步骤几乎相同。但我不知道为什么这是错误的,以及如何处理它。帮帮我,非常感谢!

2 回复 | 直到 8 年前

Vladimir Bystricky 8 年前

第一个代码 调整大小 形象第二个代码庄稼形象这是两个不同的动作,第一个动作将整个图像缩放到新的大小,第二个动作将图像的一部分剪切,使其大小相等 224 x 224

更新:

请尝试以下代码:

def test_one_image(path):
    with tf.Graph().as_default():
        image_plt = Image.open(path)
        image_tensor = tf.image.decode_jpeg(tf.read_file(path), channels=3)
        image_tensor = tf.expand_dims(image_tensor, [0])
        image_tensor = tf.image.resize_images(image_tensor, [224, 224])
        logit = vgg(image_tensor)
        logit = tf.nn.softmax(logit)
        logs_train_dir = 'D:\\Demo\\ckpt'
        saver = tf.train.Saver(tf.global_variables())
        with tf.Session() as sess:
            ckpt = tf.train.get_checkpoint_state(logs_train_dir)
            saver.restore(sess, ckpt.model_checkpoint_path)              
            prediction = sess.run(logit)
            print(prediction)

Ericccccc 8 年前

当我使用此代码时,发生了错误。结果再次为[0.33333334 0.33333334 0.3333333 4]。

def test_one_image(path):
    with tf.Graph().as_default():
        image_plt = Image.open(path)
        image_tensor = tf.image.decode_jpeg(tf.read_file(path), channels=3)
        image_tensor = tf.image.per_image_standardization(image_tensor)
        image_tensor = tf.reshape(image_tensor, [1, 224, 224, 3])
        x = tf.placeholder(tf.float32, shape=[1, 224, 224, 3])
        logit = vgg(x)
        logit = tf.nn.softmax(logit)
        logs_train_dir = 'D:\\Demo\\ckpt'
        saver = tf.train.Saver(tf.global_variables())
        with tf.Session() as sess:
            ckpt = tf.train.get_checkpoint_state(logs_train_dir)
            saver.restore(sess, ckpt.model_checkpoint_path)              
            prediction = sess.run(logit, feed_dict={x: image_tensor.eval()})
            print(prediction)