Vladimir Robles-Bykbaev

Clasificación de imágenes usando características globales y modelos de Máxima Entropía.

The recovery image process is an important task in the area of computer science. The applied tech... more The recovery image process is an important task in the area of computer science. The applied techniques for the well known search engines like Google, Yahoo, Altavista and others are based on searching the labels of the images. Such methods are prone to errors, since about the picture you can find words that have no connection with the contents searched. For all this, it is essential to have methods that make a real search-level image content. In this paper we apply feature extraction technique based on color histograms, in order to obtain images similar to a given random sample. Similarly, it is proposed to classify images using a model based on Maximum Entropy. This model creates a probability distribution based on the characteristics of the image, which is the color histogram bins.

Uploads

Talks by Vladimir Robles-Bykbaev

Log In