Sensitivity Analysis of Hopfield Neural Network in Classifying Natural RGB Color Images
Sammouda, Rachid . 2009
This paper presents a study of the sensitivity analysis of the artificial Hopfield Neural Network (HNN) when segmenting natural color images. The color distinction or vision system relies on two step process which, first classifies the different regions in the scene into a given number of clusters, and then assigns to each cluster a color that is likely to one of its corresponding region in the raw image. The classification process is performed using the minimization of an energy function typically the Sum of Squared Errors (SSE). The optimization process is found sensitive to the step taken by the network in its way to the global minimum. The color assignment to the clusters is performed based on combination of information from the color palette used in the raw image and the last distribution of the pixels among clusters. Applying the system to a gold standard color image, the results show that HNN natural color segmentation accuracy can be significantly improved if we control its step size when modifying its weights between its neurons each iteration. The color matching process shows a lot of consistency when tested with natural color images as shown in the results presented here.
نوع عمل المنشور
Research
رقم المجلد
8
رقم الانشاء
No- 9
مجلة/صحيفة
http://www.wseas.us/e-library/transactions/computers/2009/31-466.pdf
الصفحات
1514-1521
Presents contributions to improve a previously published approach for the segmentation of magnetic resonance images of the human brain, based on an unsupervised Hopfield neural network.
The paper presents a method for automatic segmentation of sputum cells color images, to develop an efficient algorithm for lung cancer diagnosis based on a Hopfield neural network.