Systems and methods for real-time horizon detection in images

摘要

Methods for detecting a horizon in an image with a plurality of pixels can include the step of blurring the image with a noise filter, then dividing the image into an M×N matrix of sub-blocks S. For each sub-block S, horizon features can be coarse-extracted by defining an r-dimensional vector having P feature values for each sub-block S and clustering each r-dimensional vectors into two clusters using a k-means statistical analysis. The corresponding sub-blocks S corresponding to the two clusters can be masked with a binary mask. The methods can further include the step of fine-extracting the horizon features at a pixel level for each sub-block Si, j and sub-block Si−1, j when the binary mask changes value from sub-block Si−1 j to said sub-block Si, j, for i=1 to M and j=1 to N.

主权项

1. A method for detecting a horizon in an image, said method comprising the steps of:
A) blurring said image; B) dividing said image into an M×N matrix of sub-blocks Sij, for i=1 to M and J=1 to N, each said sub-block Sij having a plurality of pixels; C) coarse-extracting horizon features at a said sub-block Sij granularity level by defining an r-dimensional vector having r feature values for each of said sub-blocks Sij; D) clustering each of said r-dimensional vectors into two clusters using said feature values; E) masking each of said sub-blocks Sij at a said sub-block Sij granularity level corresponding to said r-dimensional vectors using a binary mask according to the results of said step D); E1) selecting a sub-blocks Sij and sub-blocks Si+1, j from said step E) when sub-block Sij has a different mask value from the sub-block Si+1, j, step E1) being accomplished for i=1 to M−1 and J=1 to N; and, F) fine-extracting said horizon features at a pixel level, said fine-extracting step being accomplished only on the sub-blocks Sij from said step E1.