Pattern-Recognition

Projects from Pattern Recognition Course - Persian Gulf University

Introduction

Welcome to the Pattern Recognition project repository! This project focuses on implementing various pattern recognition methods on the Indian Pine Hyperspectral Imaging (HSI) dataset. The implemented methods include Linear Discriminant Analysis (LDA), Maximum A Posteriori (MAP), Principal Component Analysis (PCA), Mahalanobis Distance, Minimum Distance, Cointoss Markov Modeling, and Viterbi Algorithm.

Dataset

Indian Pine HSI Dataset

The Indian Pine dataset is a widely used hyperspectral dataset captured by the AVIRIS sensor over the Indian Pines test site in Indiana, USA. It consists of 224 spectral bands and covers an area of agricultural fields, forests, and bare soil.

Implemented Methods

1. Linear Discriminant Analysis (LDA)

Linear Discriminant Analysis is a dimensionality reduction technique used for feature extraction and classification. The goal is to find a linear combination of features that characterizes or separates classes.

2. Maximum A Posteriori (MAP)

Maximum A Posteriori is a statistical method that combines prior knowledge and likelihood to estimate the parameters of a model. In the context of pattern recognition, it is often used for classification tasks.

3. Principal Component Analysis (PCA)

Principal Component Analysis is a dimensionality reduction technique that transforms the data into a new coordinate system, capturing the most significant variance in the data.

4. Mahalanobis Distance

Mahalanobis Distance is a measure of the distance between a point and a distribution, considering the correlation between variables. It is commonly used in statistical pattern recognition.

5. Minimum Distance

Minimum Distance is a simple classification method based on the distance between data points and class centroids. It assigns a data point to the class with the nearest centroid.

6. Cointoss Markov Modeling

Cointoss Markov Modeling is a probabilistic model used for time-series data. It is applied here for pattern recognition, capturing the sequential dependencies in the data.

7. Viterbi Algorithm

The Viterbi Algorithm is a dynamic programming algorithm used to find the most likely sequence of hidden states in a hidden Markov model. It is applied in this project for sequence labeling and classification.