This section provides a description of the functions that are commonly applied in image analysis applications.
Boundary and perimeter computation functions are provided as commonly-used structural operators in vision applications.
The morphological operator functions are used to perform basic dilation and erosion operations on binary images. Dilation and erosion are the fundamental building blocks for higher-level morphological operations such as Opening and Closing. These functions are useful in machine vision and medical imaging applications.
Edge detection is a commonly-used operation in vision systems. Many algorithms exist for edge detection. One of the most commonly used methods for edge detection is the Sobel algorithm. The following functions are optimized implementation of the Sobel operator with different mask sizes and data formats.
The histogram routines provide the ability to generate an image histogram. An image histogram is basically a count of the intensity levels (or some other statistic) in an image. For example, for a grayscale image with 8-bit pixel intensity values, the histogram will consist of 256 bins corresponding to the 256 possible pixel intensities. Each bin contains a count of pixels within the image that have the particular intensity value. Histogram processing (such as histogram equalization or modification) is used in areas such as vision systems and image/video content generation systems. The 16-bit version can operate on images with data resolution from 1 to 16 bits.
Different forms of image thresholding operations are used for various purposes in image/video processing systems. For example, one form of thresholding may be used to convert grayscale image data to binary image data for input to binary morphological processing. Another form of thresholding may be used to clip image data levels into a desired range, and yet another form of thresholding may be used to zero out low-level perturbations in image data due to sensor noise. Thresholding is also found in machine vision applications to provide a simple means for image segmentation.
Copyright 2012, Texas Instruments Incorporated