1.代码地址：https://github.com/balancap/SSD-Tensorflow，下载该代码到本地
2.解压ssd_300_vgg.ckpt.zip 到checkpoint文件夹下
3.测试一下看看，在notebooks中创建demo_test.py,其实就是复制ssd_notebook.ipynb中的代码，该py文件是完成对于单张图片的测试，对Jupyter不熟，就自己改了，感觉这样要方便一些。
import os
import math
import random
import numpy as np
import tensorflow as tf
import cv2
slim = tf.contrib.slim
import matplotlib.pyplot as plt
import matplotlib.image as mpimg
import sys
sys.path.append('../')
from nets import ssd_vgg_300, ssd_common, np_methods
from preprocessing import ssd_vgg_preprocessing
from notebooks import visualization
# TensorFlow session: grow memory when needed. TF, DO NOT USE ALL MY GPU MEMORY!!!
gpu_options = tf.GPUOptions(allow_growth=True)
config = tf.ConfigProto(log_device_placement=False, gpu_options=gpu_options)
isess = tf.InteractiveSession(config=config)
# Input placeholder.
net_shape = (300, 300)
data_format = 'NHWC'
img_input = tf.placeholder(tf.uint8, shape=(None, None, 3))
# Evaluation pre-processing: resize to SSD net shape.
image_pre, labels_pre, bboxes_pre, bbox_img = ssd_vgg_preprocessing.preprocess_for_eval(
img_input, None, None, net_shape, data_format, resize=ssd_vgg_preprocessing.Resize.WARP_RESIZE)
image_4d = tf.expand_dims(image_pre, 0)
# Define the SSD model.
reuse = True if 'ssd_net' in locals() else None
ssd_net = ssd_vgg_300.SSDNet()
with slim.arg_scope(ssd_net.arg_scope(data_format=data_format)):
predictions, localisations, _, _ = ssd_net.net(image_4d, is_training=False, reuse=reuse)
# Restore SSD model.
ckpt_filename = '../checkpoints/ssd_300_vgg.ckpt'
# ckpt_filename = '../checkpoints/VGG_VOC0712_SSD_300x300_ft_iter_120000.ckpt'
isess.run(tf.global_variables_initializer())
saver = tf.train.Saver()
saver.restore(isess, ckpt_filename)
# SSD default anchor boxes.
ssd_anchors = ssd_net.anchors(net_shape)
# Main image processing routine.
def process_image(img, select_threshold=0.5, nms_threshold=.45, net_shape=(300, 300)):
# Run SSD network.
rimg, rpredictions, rlocalisations, rbbox_img = isess.run([image_4d, predictions, localisations, bbox_img],
feed_dict={img_input: img})
# Get classes and bboxes from the net outputs.
rclasses, rscores, rbboxes = np_methods.ssd_bboxes_select(
rpredictions, rlocalisations, ssd_anchors,
select_threshold=select_threshold, img_shape=net_shape, num_classes=21, decode=True)
rbboxes = np_methods.bboxes_clip(rbbox_img, rbboxes)
rclasses, rscores, rbboxes = np_methods.bboxes_sort(rclasses, rscores, rbboxes, top_k=400)
rclasses, rscores, rbboxes = np_methods.bboxes_nms(rclasses, rscores, rbboxes, nms_threshold=nms_threshold)
# Resize bboxes to original image shape. Note: useless for Resize.WARP!
rbboxes = np_methods.bboxes_resize(rbbox_img, rbboxes)
return rclasses, rscores, rbboxes
# Test on some demo image and visualize output.
#测试的文件夹
path = '../demo/'
image_names = sorted(os.listdir(path))
#文件夹中的第几张图，-1代表最后一张
img = mpimg.imread(path + image_names[-1])
rclasses, rscores, rbboxes = process_image(img)
# visualization.bboxes_draw_on_img(img, rclasses, rscores, rbboxes, visualization.colors_plasma)
visualization.plt_bboxes(img, rclasses, rscores, rbboxes)
 |
 |
 |
4.将自己的数据集做成VOC2007格式放在该工程下面
 |
 |
 |
 |
5. 修改datasets文件夹中pascalvoc_common.py文件，将训练类修改别成自己的
#原始的
# VOC_LABELS = {
# 'none': (0, 'Background'),
# 'aeroplane': (1, 'Vehicle'),
# 'bicycle': (2, 'Vehicle'),
# 'bird': (3, 'Animal'),
# 'boat': (4, 'Vehicle'),
# 'bottle': (5, 'Indoor'),
# 'bus': (6, 'Vehicle'),
# 'car': (7, 'Vehicle'),
# 'cat': (8, 'Animal'),
# 'chair': (9, 'Indoor'),
# 'cow': (10, 'Animal'),
# 'diningtable': (11, 'Indoor'),
# 'dog': (12, 'Animal'),
# 'horse': (13, 'Animal'),
# 'motorbike': (14, 'Vehicle'),
# 'person': (15, 'Person'),
# 'pottedplant': (16, 'Indoor'),
# 'sheep': (17, 'Animal'),
# 'sofa': (18, 'Indoor'),
# 'train': (19, 'Vehicle'),
# 'tvmonitor': (20, 'Indoor'),
# }
#修改后的
VOC_LABELS = {
'none': (0, 'Background'),
'pantograph':(1,'Vehicle'),
}
6. 将图像数据转换为tfrecods格式，修改datasets文件夹中的pascalvoc_to_tfrecords.py文件，然后更改文件的83行读取方式为’rb‘，如果你的文件不是.jpg格式，也可以修改图片的类型。
此外， 修改67行，可以修改几张图片转为一个tfrecords
7.运行tf_convert_data.py文件，但是需要传给它一些参数：
| linux |
在SSD-Tensorflow-master文件夹下创建tf_conver_data.sh,文件写入内容如下： DATASET_DIR=./VOC2007/ #VOC数据保存的文件夹（VOC的目录格式未改变） |
windows +pychram |
配置pycharm-->run-->Edit Configuration Script parameters中写入：--dataset_name=pascalvoc --dataset_dir=./VOC2007/ --output_name=voc_2007_train --output_dir=./tfrecords_ 运行tf_convert_data.py文件 |
| 生成tfrecords文件过程中你会看到 | 生成tfrecords文件完毕后你会看到 |
 |
 |
8.训练模型train_ssd_network.py文件中修改
train_ssd_network.py文件中网络参数配置，若需要改，在此文件中进行修改，如：
其他需要修改的地方
| a. nets/ssd_vgg_300.py (因为使用此网络结构) ，修改87 和88行的类别 |  |
||
| b. train_ssd_network.py,修改类别120行，GPU占用量，学习率，batch_size等 |  |
||
| c eval_ssd_network.py 修改类别，66行 |  |
||
| d. datasets/pascalvoc_2007.py 根据自己的训练数据修改整个文件 |
|
9.通过加载预训练好的vgg16模型,训练网络
下载预训练好的vgg16模型，解压放入checkpoint文件中，如果找不到vgg_16.ckpt文件，可以在下面的链接中点击下载。
链接：https://pan.baidu.com/s/1diWbdJdjVbB3AWN99406nA 密码：ge3x
按照之前的方式，同样，如果你是linux用户，你可以新建一个.sh文件，文件里写入
DATASET_DIR=./tfrecords_/
TRAIN_DIR=./train_model/
CHECKPOINT_PATH=./checkpoints/vgg_16.ckpt
python3 ./train_ssd_network.py \
--train_dir=./train_model/ \ #训练生成模型的存放路径
--dataset_dir=./tfrecords_/ \ #数据存放路径
--dataset_name=pascalvoc_2007 \ #数据名的前缀
--dataset_split_name=train \
--model_name=ssd_300_vgg \ #加载的模型的名字
--checkpoint_path=./checkpoints/vgg_16.ckpt \ #所加载模型的路径
--checkpoint_model_scope=vgg_16 \ #所加载模型里面的作用域名
--checkpoint_exclude_scopes=ssd_300_vgg/conv6,ssd_300_vgg/conv7,ssd_300_vgg/block8,ssd_300_vgg/block9,ssd_300_vgg/block10,ssd_300_vgg/block11,ssd_300_vgg/block4_box,ssd_300_vgg/block7_box,ssd_300_vgg/block8_box,ssd_300_vgg/block9_box,ssd_300_vgg/block10_box,ssd_300_vgg/block11_box \
--trainable_scopes=ssd_300_vgg/conv6,ssd_300_vgg/conv7,ssd_300_vgg/block8,ssd_300_vgg/block9,ssd_300_vgg/block10,ssd_300_vgg/block11,ssd_300_vgg/block4_box,ssd_300_vgg/block7_box,ssd_300_vgg/block8_box,ssd_300_vgg/block9_box,ssd_300_vgg/block10_box,ssd_300_vgg/block11_box \
--save_summaries_secs=60 \ #每60s保存一下日志
--save_interval_secs=600 \ #每600s保存一下模型
--weight_decay=0.0005 \ #正则化的权值衰减的系数
--optimizer=adam \ #选取的最优化函数
--learning_rate=0.001 \ #学习率
--learning_rate_decay_factor=0.94 \ #学习率的衰减因子
--batch_size=24 \
--gpu_memory_fraction=0.9 #指定占用gpu内存的百分比
如果你是windows+pycharm中运行，除了在上述的run中Edit Configuration配置，你还可以打开Terminal，在这里运行代码，输入即可
python ./train_ssd_network.py --train_dir=./train_model/ --dataset_dir=./tfrecords_/ --dataset_name=pascalvoc_2007 --dataset_split_name=train --model_name=ssd_300_vgg --checkpoint_path=./checkpoints/ --checkpoint_model_scope=vgg_16 --checkpoint_exclude_scopes=ssd_300_vgg/conv6,ssd_300_vgg/conv7,ssd_300_vgg/block8,ssd_300_vgg/block9,ssd_300_vgg/block10,ssd_300_vgg/block11,ssd_300_vgg/block4_box,ssd_300_vgg/block7_box,ssd_300_vgg/block8_box,ssd_300_vgg/block9_box,ssd_300_vgg/block10_box,ssd_300_vgg/block11_box --trainable_scopes=ssd_300_vgg/conv6,ssd_300_vgg/conv7,ssd_300_vgg/block8,ssd_300_vgg/block9,ssd_300_vgg/block10,ssd_300_vgg/block11,ssd_300_vgg/block4_box,ssd_300_vgg/block7_box,ssd_300_vgg/block8_box,ssd_300_vgg/block9_box,ssd_300_vgg/block10_box,ssd_300_vgg/block11_box --save_summaries_secs=60 --save_interval_secs=600 --weight_decay=0.0005 --optimizer=adam --learning_rate=0.001 --learning_rate_decay_factor=0.94 --batch_size=24 --gpu_memory_fraction=0.9