Using TensorFlow to implement camera image processing under C #

The TensorFlow self-contained example already contains a camera image recognition under Android and iOS, which complements a Windows-operated camera using the Aforge Library (www.aforgenet.com).

The code is downloaded here and compiled with VS2017.

Http://files.cnblogs.com/files/autosoftdev/TensorFlowCamera.zip

Run-time, from here

https://www.nuget.org/api/v2/package/TensorFlowSharp/1.2.2

Download and unzip to remove the Libtensorflow.dll.

From here

Https://storage.googleapis.com/download.tensorflow.org/models/inception5h.zip

Download the model.

And put Libtensorflow.dll, TENSORFLOW_INCEPTION_GRAPH.PB, imagenet_comp_graph_label_strings.txt these 3 files into the compiled exe sibling directory.

Run to see the camera screen and display the recognition results in real time. All the information of the program comes from the network collection and collation, only for learning.

Resources:

Http://www.cnblogs.com/Leo_wl/p/6897649.html
Https://github.com/migueldeicaza/TensorFlowSharp

Using TensorFlow to implement camera image processing under C #