TensorFlow Inception model is used to perform classification on number of images present in a directory.
To run this script you need to have Python and TensorFlow installed on your system
Once you have installed above prerequisites place all JPEG or JPG image files in img folder and run the script
python mul_classify_image.py
It will generate a result.txt file which consits of filenames, prediction labels and scores.
Note: sample output for images present in img folder is shown below
Filename: image_5.jpg
TensorFlow - Image Classification:
tiger, Panthera tigris - 78.537%
tiger cat - 9.822%
jaguar, panther, Panthera onca, Felis onca - 0.753%
lynx, catamount - 0.223%
leopard, Panthera pardus - 0.158%
Filename: image_1.jpg
TensorFlow - Image Classification:
meerkat, mierkat - 93.775%
mongoose - 1.971%
Windsor tie - 0.236%
black-footed ferret, ferret, Mustela nigripes - 0.060%
marmot - 0.031%
Filename: image_3.jpg
TensorFlow - Image Classification:
red fox, Vulpes vulpes - 91.107%
kit fox, Vulpes macrotis - 2.673%
grey fox, gray fox, Urocyon cinereoargenteus - 0.373%
dhole, Cuon alpinus - 0.177%
red wolf, maned wolf, Canis rufus, Canis niger - 0.167%
Filename: image_2.jpg
TensorFlow - Image Classification:
hyena, hyaena - 91.135%
African hunting dog, hyena dog, Cape hunting dog, Lycaon pictus - 1.869%
dhole, Cuon alpinus - 0.138%
dugong, Dugong dugon - 0.099%
custard apple - 0.035%
Filename: image_4.jpeg
TensorFlow - Image Classification:
tiger, Panthera tigris - 81.967%
tiger cat - 12.465%
zebra - 0.124%
jaguar, panther, Panthera onca, Felis onca - 0.074%
lynx, catamount - 0.055%
All files present in above folder are developed by TensorFlow authors. I have just added few lines of code and modified its classification script.