树莓派3B+ tensorflow目标检测API SSD

整个工程已经上传到Github上了

https://github.com/xyc2690/Raspberry_ObjectDetection_Camera

环境准备：摄像头，OpenCV

首先在树莓派上安装最新的TensorFlow

https://github.com/lhelontra/tensorflow-on-arm/releases

下载SSDmobile模型

http://download.tensorflow.org/models/object_detection/ssdlite_mobilenet_v2_coco_2018_05_09.tar.gz

安装OpenCV，

扫描二维码关注公众号，回复： 1704523 查看本文章

只用python的话可以直接 pip install opencv-python

或者下载最新完整的OpenCV手动编译。

然后直接测试摄像头

python opencv_camera.py

修改参数，尝试tensorflow提供的ObjectDetectionAPI吧

TEST_CAM_ONLY = False