README
¶
carve
Go implementation of Seam Carving algorithm. Seam carving allows for an image to be resized without simply cropping or scaling the original image. Detailed walk through of the algorithm and code can be found on my website. Algorithm is also detailed on Wikipedia, the original paper.
Installation
$ go get https://github.com/aaparella/carve
Usage
resized, err := carve.ReduceHeight(img, 50)
if err != nil {
log.Fatal(err)
}
Todo
- Improve performance
- Reduce heap allocations
- Caching?
Documentation
¶
Overview ¶
Package carve implements the seam carving algorithm for dyanmic image resizing.
Index ¶
- func GenerateCostMatrix(im image.Image) [][]float64
- func GenerateEnergyMap(im image.Image) image.Image
- func ReduceHeight(im image.Image, n int) (image.Image, error)
- func ReduceWidth(im image.Image, n int) (image.Image, error)
- func RemoveSeam(im image.Image, seam Seam) image.Image
- type Point
- type Seam
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
func GenerateCostMatrix ¶
GenerateCostMatrix creates a matrix indicating the cumulative energy of the lowest cost seam from the left of the image to each pixel.
mat[x][y] is the cumulative energy of the seam that runs from the left of the image to the pixel at column x, row y.
func GenerateEnergyMap ¶
GenerateEnergyMap applies grayscale and sobel filters to the input image to create an energy map.
func ReduceHeight ¶
ReduceHeight uses seam carving to reduce height of given image by n pixels.
func ReduceWidth ¶
ReduceWidth uses seam carving to reduce width of given image by n pixels.
Types ¶
type Seam ¶
type Seam []Point
Seam defines a sequence of pixels through an image to be removed.
func FindLowestCostSeam ¶
FindLowestCostSeam uses a cumulative cost matrix to identify the seam with the lowest total cumulative energy.
func GenerateSeam ¶
GenerateSeam returns the optimal horizontal seam for removal.
Source Files