<![CDATA[This writing investigates the classification of images taken using a smartphone. Due to the large number of photos and the large number of photo categories, it is necessary to automatically categorize these photos. Photos are classified using two different approaches. The first method uses Hidden Markov Model (HMM) and the second technique employs Siamese Network from Convolutional Neural Network (CNN) architecture. The same data are used for training and testing for both models. For HMM we use Discrete Cosine Transform (DCT) to extract salient features of images. The number of training examples is very small compared to the test set. Here we carried out few-shot classification method. For recognition of the HMM, Viterbi algorithm is applied. Performances of both procedures were measured. For only 109 test samples HMM achieve 98% accuracy, while twin network achieves 90%. The use of HMM has advantage over Siamese in term of faster computation. HMM opens the opportunity of the smartphone with low computation capability to categorize photos automatically.]]>
Copyrights © 2025
