Minimum distance. Gaussian maximum likelihood . Parallelepiped

Minimum distance to means classification is a supervised classification technique in remote sensing that works by dividing the data into a number of classes based on the mean value of each class. The algorithm works as follows:

First, the mean value of each class is calculated. This is done by taking the average of all the data points in each class.

Next, for each data point, the distance to the mean of each class is calculated. This is done using a distance metric, such as Euclidean distance.

The data point is then assigned to the class with the minimum distance to the mean.

This process is repeated for all data points in the dataset.

Minimum distance to means classification is simple and easy to implement, but it can be sensitive to noise and outliers in the data. It is generally not as accurate as more complex classification algorithms, such as support vector machines or neural networks.

Gaussian maximum likelihood classification is a method of image analysis in remote sensing that involves estimating the probability density function (PDF) of each class in the image and then classifying each pixel based on the class with the highest PDF value.

The PDF of a class is a statistical model that describes the distribution of pixel values within that class. In the case of Gaussian maximum likelihood classification, the PDF is assumed to be a Gaussian, or normal, distribution. This means that the pixel values within the class are assumed to be normally distributed around a mean value, with a certain standard deviation.

To classify each pixel, the PDFs of all classes are calculated using the mean and standard deviation values estimated from the training data. The class with the highest PDF value is then assigned to the pixel.

One advantage of Gaussian maximum likelihood classification is that it can handle continuous variables, such as spectral reflectance values, which can be difficult to classify using other methods. It is also relatively simple to implement and can be easily modified to incorporate additional features or constraints.

However, Gaussian maximum likelihood classification has some limitations. It assumes that the classes are normally distributed, which may not always be the case in real-world data. It is also sensitive to the presence of mixed pixels, or pixels that contain multiple types of land cover.

In remote sensing, a parallelepiped is a three-dimensional model used to classify spectral data obtained from a remote sensor. It is a parallelogram with opposite sides parallel, and all its faces are parallelograms.

Parallelepiped classification is a method of image analysis that involves dividing the image data into a set of smaller parallelepipeds, or "bins," and then assigning a class label to each bin based on the characteristics of the pixels within it. This can be done using various techniques, such as k-means clustering or decision tree analysis.

One advantage of parallelepiped classification is that it can be used to analyze large volumes of data quickly, as the bins can be processed in parallel. It is also relatively simple to implement and can be easily modified to incorporate additional features or constraints.

However, parallelepiped classification has some limitations. It can be sensitive to the size and orientation of the bins, and the choice of bin size and orientation can significantly affect the accuracy of the classification. It can also be sensitive to the presence of mixed pixels, or pixels that contain multiple types of land cover.

Comments

RADIOMETRIC CORRECTION

Radiometric correction is the process of removing sensor and environmental errors from satellite images so that the measured brightness values (Digital Numbers or DNs) truly represent the Earth's surface reflectance or radiance. In other words, it corrects for sensor defects, illumination differences, and atmospheric effects. 1. Detector Response Calibration Satellite sensors use multiple detectors to scan the Earth's surface. Sometimes, each detector responds slightly differently, causing distortions in the image. Calibration adjusts all detectors to respond uniformly. This includes: (a) De-Striping Problem: Sometimes images show light and dark vertical or horizontal stripes (banding). Caused by one or more detectors drifting away from their normal calibration — they record higher or lower values than others. Common in early Landsat MSS data. Effect: Every few lines (e.g., every 6th line) appear consistently brighter or darker. Soluti...

Geometric Correction

When satellite or aerial images are captured, they often contain distortions (errors in shape, scale, or position) caused by many factors — like Earth's curvature, satellite motion, terrain height (relief), or the Earth's rotation . These distortions make the image not properly aligned with real-world coordinates (latitude and longitude). 👉 Geometric correction is the process of removing these distortions so that every pixel in the image correctly represents its location on the Earth's surface. After geometric correction, the image becomes geographically referenced and can be used with maps and GIS data. Types 1. Systematic Correction Systematic errors are predictable and can be modeled mathematically. They occur due to the geometry and movement of the satellite sensor or the Earth. Common systematic distortions: Scan skew – due to the motion of the sensor as it scans the Earth. Mirror velocity variation – scanning mirror moves at a va...

Atmospheric Correction

It is the process of removing the influence of the atmosphere from remotely sensed images so that the data accurately represent the true reflectance of Earth's surface . When a satellite sensor captures an image, the radiation reaching the sensor is affected by gases, water vapor, aerosols, and dust in the atmosphere. These factors scatter and absorb light, changing the brightness and color of the features seen in the image. Although these atmospheric effects are part of the recorded signal, they can distort surface reflectance values , especially when images are compared across different dates or sensors . Therefore, corrections are necessary to make data consistent and physically meaningful. 🔹 Why Do We Need Atmospheric Correction? To retrieve true surface reflectance – It separates the surface signal from atmospheric influence. To ensure comparability – Enables comparing images from different times, seasons, or sensors. To improve visual quality – Remo...

Supervised Classification

In the context of Remote Sensing (RS) and Digital Image Processing (DIP) , supervised classification is the process where an analyst defines "training sites" (Areas of Interest or ROIs) representing known land cover classes (e.g., Water, Forest, Urban). The computer then uses these training samples to teach an algorithm how to classify the rest of the image pixels. The algorithms used to classify these pixels are generally divided into two broad categories: Parametric and Nonparametric decision rules. Parametric Decision Rules These algorithms assume that the pixel values in the training data follow a specific statistical distribution—almost always the Gaussian (Normal) distribution (the "Bell Curve"). Key Concept: They model the data using statistical parameters: the Mean vector ( $\mu$ ) and the Covariance matrix ( $\Sigma$ ) . Analogy: Imagine trying to fit a smooth hill over your data points. If a new point lands high up on the hill, it belongs to that cl...

Supervised Classification

Image Classification in Remote Sensing Image classification in remote sensing involves categorizing pixels in an image into thematic classes to produce a map. This process is essential for land use and land cover mapping, environmental studies, and resource management. The two primary methods for classification are Supervised and Unsupervised Classification . Here's a breakdown of these methods and the key stages of image classification. 1. Types of Classification Supervised Classification In supervised classification, the analyst manually defines classes of interest (known as information classes ), such as "water," "urban," or "vegetation," and identifies training areas —sections of the image that are representative of these classes. Using these training areas, the algorithm learns the spectral characteristics of each class and applies them to classify the entire image. When to Use Supervised Classification: - You have prior knowledge about the c...

Vineesh V, Geography

Search This Blog