Implements filter bank for encoding local binary patterns. More...

Public Member Functions
	drwnLBPFilterBank (bool b8Neighbourbood=false)
	construct an LBP filterbank

size_t	numFilters () const
	returns the number of response channels

void	filter (const cv::Mat &img, std::vector< cv::Mat > &response) const
	Filtering function. The caller must provide a vector of CV32F destination matrices (or empty). The source image should be a 3-channel RGB color or 1-channel greyscale image.

void	regionFeatures (const cv::Mat &regions, const std::vector< cv::Mat > &response, vector< vector< double > > &features)
	Compute LBP histograms over given superpixel regions. The `response` input comes from running the `filter` member function. Features are output in `features`.

void	regionFeatures (const drwnSuperpixelContainer &regions, const std::vector< cv::Mat > &response, vector< vector< double > > &features)
	Compute LBP histograms over given superpixel regions. The `response` input comes from running the `filter` member function. Features are output in `features`.

Protected Attributes
bool	_b8Neighbourhood
	true of 8-connected neighbourhood, otherwise < a 4-connected is used

Detailed Description

Implements filter bank for encoding local binary patterns.

Aggregating the responses over regions and normalizing give the standard LBP feature.

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

Public Member Functions