textboxes中数据集的产生到使用

tfrecord的生产

针对的数据集：SythTexts datasets(41G)合成数据集

1产生tfrecord的所需的变量

def _processing_image(wordbb, imname,coder):
	#wordbb = tf.cast(wordbb, tf.float32)
	image_data = tf.gfile.GFile(FLAGS.datasets_jpgfile + imname, 'r').read()
	#获取图片，这里image_data是一个tensor
	
	image = coder.decode_jpeg(image_data)
	#得到图片的数据 rgb格式 coder.decode_jpeg =sess.run(tf.image.decode_jpeg(....),feed_dic=image_data)
	shape = image.shape #[h,w,c]
	if(len(wordbb.shape) < 3 ):
		numofbox = 1
	else:
		numofbox = wordbb.shape[2]
	bbox = []
	[xmin, ymin]= np.min(wordbb,1)
	[xmax, ymax] = np.max(wordbb,1)
	xmin = np.maximum(xmin*1./shape[1], 0.0)
	ymin = np.maximum(ymin*1./shape[0], 0.0)
	xmax = np.minimum(xmax*1./shape[1], 1.0)
	ymax = np.minimum(ymax*1./shape[0], 1.0)
	if numofbox > 1:
		bbox = [[ymin[i],xmin[i],ymax[i],xmax[i]] for i in range(numofbox)] 
	if numofbox == 1:
		bbox = [[ymin,xmin,ymax,xmax]]


	label = [1 for i in range(numofbox)]#有多少个bbox，就有多少个1标签
	shape = list(shape)
	return image_data, shape, bbox, label, imname

这里所需的变量有 image_data，shape，bbox，label，imname

2生成.tfrecord的编辑对象

def _convert_to_example(image_data, shape, bbox, label,imname):
	nbbox = np.array(bbox)
	ymin = list(nbbox[:, 0])
	xmin = list(nbbox[:, 1])
	ymax = list(nbbox[:, 2])
	xmax = list(nbbox[:, 3])

	#print 'shape: {}, height:{}, width:{}'.format(shape,shape[0],shape[1])
	example = tf.train.Example(features=tf.train.Features(feature={
			'image/height': int64_feature(shape[0]),
			'image/width': int64_feature(shape[1]),
			'image/channels': int64_feature(shape[2]),
			'image/shape': int64_feature(shape),
			'image/object/bbox/ymin': float_feature(ymin),
			'image/object/bbox/xmin': float_feature(xmin),
			'image/object/bbox/ymax': float_feature(ymax),
			'image/object/bbox/xmax': float_feature(xmax),
			'image/object/bbox/label': int64_feature(label),
			'image/format': bytes_feature('jpeg'),
			'image/encoded': bytes_feature(image_data),
			'image/name': bytes_feature(imname.tostring()),
			}))
	return example

3 写入tfrecord

tf_filename = str(i+1) + '.tfrecord'
tfrecord_writer = tf.python_io.TFRecordWriter(FLAGS.path_save + tf_filename)
tfrecord_writer.write(example.SerializeToString()) 

4使用tfrecord

import tensorflow as tf
import os
slim = tf.contrib.slim




ITEMS_TO_DESCRIPTIONS = {
    'image': 'slim.tfexample_decoder.Image',
    'shape': 'shape',
    'height': 'height',
    'width': 'width',
    'object/bbox': 'box',
    'object/label': 'label'
}
SPLITS_TO_SIZES = {
    'train': 858750,
}
NUM_CLASSES = 2



def get_datasets(data_dir,file_pattern = '*.tfrecord'):
    file_patterns = os.path.join(data_dir, file_pattern)
    print 'file_path: {}'.format(file_patterns)
    reader = tf.TFRecordReader
    keys_to_features = {
        'image/height': tf.FixedLenFeature([1], tf.int64),
        'image/width': tf.FixedLenFeature([1], tf.int64),
        'image/channels': tf.FixedLenFeature([1], tf.int64),
        'image/shape': tf.FixedLenFeature([3], tf.int64),
        'image/object/bbox/ymin': tf.VarLenFeature(dtype=tf.float32),
        'image/object/bbox/xmin': tf.VarLenFeature(dtype=tf.float32),
        'image/object/bbox/ymax': tf.VarLenFeature(dtype=tf.float32),
        'image/object/bbox/xmax': tf.VarLenFeature(dtype=tf.float32),
        'image/object/bbox/label': tf.VarLenFeature(dtype=tf.int64),
        'image/format': tf.FixedLenFeature([], tf.string, default_value='jpeg'),
        'image/encoded': tf.FixedLenFeature([], tf.string, default_value=''),
        'image/name': tf.VarLenFeature(dtype = tf.string),
    }

    items_to_handlers = {
        'image': slim.tfexample_decoder.Image('image/encoded', 'image/format'),
        #'image': slim.tfexample_decoder.Tensor('image/encoded'),
        'shape': slim.tfexample_decoder.Tensor('image/shape'),
        'height': slim.tfexample_decoder.Tensor('image/height'),
        'width': slim.tfexample_decoder.Tensor('image/width'),
        'object/bbox': slim.tfexample_decoder.BoundingBox(
                ['ymin', 'xmin', 'ymax', 'xmax'], 'image/object/bbox/'),
        'object/label': slim.tfexample_decoder.Tensor('image/object/bbox/label'),
        #'imaname': slim.tfexample_decoder.Tensor('image/name'),
        #'objext/txt': slim.tfexample_decoder.Tensor('image/object/bbox/label_text'),
      }

    decoder = slim.tfexample_decoder.TFExampleDecoder(
        keys_to_features, items_to_handlers)

    labels_to_names = None


    return slim.dataset.Dataset(
        data_sources=file_patterns,
        reader=reader,
        decoder=decoder,
        num_samples=SPLITS_TO_SIZES['train'],
        items_to_descriptions=ITEMS_TO_DESCRIPTIONS,
        num_classes=NUM_CLASSES,
        labels_to_names=labels_to_names)

再通过设定相应的参数，就可以得到自己想要的数据

dataset = sythtextprovider.get_datasets(dataset_dir,file_pattern = file_pattern)

provider = slim.dataset_data_provider.DatasetDataProvider(
				dataset,
				num_readers=num_readers,
				common_queue_capacity=512 * 16 + 20 * batch_size,
				common_queue_min=512 * 16,
				shuffle=shuffe)
	
[image, shape, glabels, gbboxes,height,width] = provider.get(['image', 'shape',
											 'object/label',
											 'object/bbox','height','width'])