Line characteristic LSD and descriptor sub LBD---LBD algorithm (two)

The previous section mainly introduces the LSD algorithm theory, this section mainly introduces the LBD algorithm.

1. Extracting line segments in scale space

To overcome the on-line detection of fragmentation problems and to improve the performance of large-scale changes, our detection framework consists of a scale space pyramid (n-layer image) that kills samples of the original image (with a set of scale factor and Gaussian blur). We first use the Edline algorithm to generate a set of lines per layer of scale space. Each line has a direction, which consists of a gradient of most of the edge pixels from the left side of the line to the right. They are then organized by locating the corresponding lines in the scale space. In the scale space of all the lines, they are assigned a unique ID if they are related to the same event in the image (that is, the image of the same region has the same direction), storing it in a vector called linevecs. The line detection method reduces the dimension of graph matching problem by re-organizing all the segments to extract scale space to form linevecs. The result of the final extraction is a set of linevecs such as:

As shown in 1, each Linevec may include multiple lines in the scale space. To describe a Linevec local appearance, for each line, we produce a line descriptor in each layer of the extracted segment.

2. Stripe (Band) to represent the line's support domain

The main reason for proposing this new line descriptor is to describe the local appearance of the line more effectively than MSLD and to lose the matching performance.

Description Sub-BDJ of each strip: through its nearest two adjacent rows of strip Bj-1; Bj+1 to calculate. In particular, in the top and bottom strips band B1 and BM outside the LSR , in the calculation of B1 and BM Will not be taken into account when describing the

(The line Band descriptor) LBD:

Calculation of BDJ:

K: The gradient of the pixels we accumulate in this line is expressed in the stripe BJ or its other K-line.

: Gaussian weights

The band Description Matrix (BDM):

BDJ is obtained by the mean vector of the BDMJ matrix MJ and the standard variance Sj:

The mean part and the standard variance part of LBD are normalized by their different sizes. In addition, to reduce the effects of nonlinear illumination changes, LBD each dimension is suppressed, it is less than a threshold value (experience: 0.4 is a good value). Finally, we re-standardize the constraint vectors to get the LBD of the units.

Line characteristic LSD and descriptor sub LBD---LBD algorithm (two)