IMGLIB Function Reference

Release 3.1.1.0


Contents

  1. Introduction
  2. Notational Conventions
  3. Functional Reference


Introduction

IMGLIB provides a collection of C-callable high-performance routines that can serve as key enablers for a wide range of image/video processing applications. These functions exploit the high performance capabilities of Texas Instruments DSPs.

All functions in IMGLIB have been developed for a little-endian memory model. Some may work in a big-endian memory model as well, however, their functionality is not guaranteed.


Notational Conventions

All functions within IMGLIB following the naming conventions described below.

  1. Each function begins with the prefix IMG_. This serves to prevent namespace conflicts.
  2. The prefix is followed by a functional descriptor. This descriptor provides insight into the functionality of the routine as well as mask size information when relevant.
  3. A suffix is placed after each function descriptor. The suffix describes the type of inputs the function accepts. Valid suffixes are defined below.

For each of the suffixes, d represents the data length in bits and [s] indicates that the letter s may be appended to indicate that the data is signed. All fixed point math input and output data is 32-bit. A few examples follow in the table below.

Kernel Name Input Format Mask Format
IMG_conv_5x5_i8_c16s 8-bit unsigned 16-bit signed
IMG_conv_3x3_i16_c16s 16-bit unsigned 16-bit signed
IMG_corr_gen_iq 32-bit fixed-point N/A
IMG_boundary_16s 16-bit signed N/A
IMG_dilate_bin packed binary N/A


Functional Reference

The following sections provide a detailed reference of all functions within IMGLIB. The functions are organized into the following three categories:


Copyright 2012, Texas Instruments Incorporated