Mission

Let us try to identify different objects with our GoogleNet image idenfitication system.

For this mission, you will be tasked to see the image identification system work on different pictures. We will work with the with the image identification program directly from our jupyter notebook, hence we will not be using our shell programs.

Open the gogolenet_mission python notebook:

1.googlenet_mission.ipynb
Running the cell code

Ctrl + Enter

This mission will use python library called Image. This allows us to view images within our jupyter notebook environment.
```
from IPython.display import Image
```

For our jupyter notebook to run, we must set up correct file paths.

# Evnironment variables
%env PROGRAM_PATH=/home/zeta/jetson-inference/build/aarch64/bin
%env INPUT_PATH=/home/zeta/jetson-inference/build/aarch64/bin/images
%env OUTPUT_PATH=/home/zeta/jetson-inference/build/aarch64/bin/images/test/result_1.jpg

# Python variables
program_path = '/home/zeta/jetson-inference/build/aarch64/bin/'
input_path = '/home/zeta/jetson-inference/build/aarch64/bin/images/'
output_path = '/home/zeta/jetson-inference/build/aarch64/bin/images/test/result_1.jpg'

Check all the available images in our system
```
!ls $INPUT_PATH | grep \.jpg$
```

Choose one of them and change CHANGEME string with the chosen picture name.

# Enter the image file name in image_name variable
image_name = 'CHANGEME' #example: orange_0.jpg
%env IMG_NAME = $image_name

Check if the chosen picture is valid
```
Image(filename=input_path+image_name)
```

Run the program on the specified picture

!python3 $PROGRAM_PATH/imagenet.py --network=googlenet $INPUT_PATH/$IMG_NAME $OUTPUT_PATH

Display the result
```
Image(filename=output_path)
```