python3+tensorflow实时读取rtmp的流+object

python3+tensorflow实时读取rtmp的流+object

1.执行环境

win7(64) + GeForce940M 显卡+ cuda:10.0.130 + python3.7 + object_detection

2.下载Nginx

nginx 1.7.11.3 Gryphon下载连接
解压，在conf新增配置文件 。复制f文件改名为f，编辑f新增rtmp的配置

rtmp {server {listen 8080;chunk_size 4000;application live {live on;# record first 1K of streamrecord all;record_path /tmp/av;record_max_size 1K;# append current timestamp to each flvrecord_unique on;# publish only from localhostallow publish 127.0.0.1;deny publish all;#allow play all;}}
}

启动

 -c f

3.安装ffmpeg

下载地址为ffmpeg
配置好环境变量

4. 运行代码

import subprocessimport cv2
import numpy as np
import tensorflow as tf
from object_detection.utils import label_map_util
from object_detection.utils import visualization_utils as vis_utilPATH_TO_CKPT = r"E:AIfrozen_inference_graph.pb"
PATH_TO_LABELS = r"E:AImscoco_label_map.pbtxt"NUM_CLASSES = 90label_map = label_map_util.load_labelmap(PATH_TO_LABELS)
categories = label_vert_label_map_to_categories(label_map, max_num_classes=NUM_CLASSES, use_display_name=True)
category_index = label_ate_category_index(categories)detection_graph = tf.Graph