Binned mean luminance based AGC algorithm. More...

Classes
class	Limits
	Set of limits for the algorithm. More...

struct	Params
	Collection of parameters for the algorithm. More...

struct	Result
	Collection of results of the algorithm. More...

Public Member Functions
void	setLimits (const Limits &limits)
	Set the limits for the algorithm.

Result	calculateNewEv (const Params &params)
	Calculate a new set of AGC parameters.

Detailed Description

Binned mean luminance based AGC algorithm.

This is a best-effort, low-complexity auto exposure/gain algorithm based on the 2007 paper "Automatic camera exposure control" (https://www.araa.asn.au/acra/acra2007/papers/paper84final.pdf)

Due to its simplicity, it is a better fit than AgcMeanLuminance when operating a sensor without a known gain model.

The algorithm operates by inspecting the luminance histogram of a frame, and then splitting it into a fixed number of bins. Then an average is calculated (MSV ~ mean sample value), which is compared to the target MSV. If the current MSV is too low, first the exposure time, then the analogue gain is raised. If it is too high, first the analogue gain, then the exposure time is reduced.

The documentation for this class was generated from the following files:

src/ipa/libipa/agc_msv.h
src/ipa/libipa/agc_msv.cpp

Classes

Public Member Functions

Detailed Description