Files
caire/carver.go
2018-06-12 12:59:44 +03:00

394 lines
10 KiB
Go

package caire
import (
"fmt"
"image"
"image/color"
"image/draw"
"image/jpeg"
_ "image/png"
"io/ioutil"
"log"
"math"
"os"
"os/signal"
"syscall"
"time"
pigo "github.com/esimov/pigo/core"
)
var usedSeams []UsedSeams
// TempImage temporary image file.
var TempImage = fmt.Sprintf("%d.jpg", time.Now().Unix())
// Carver is the main entry struct having as parameters the newly generated image width, height and seam points.
type Carver struct {
Width int
Height int
Points []float64
}
// UsedSeams contains the already generated seams.
type UsedSeams struct {
ActiveSeam []ActiveSeam
}
// ActiveSeam contains the current seam position and color.
type ActiveSeam struct {
Seam
Pix color.Color
}
// Seam struct contains the seam pixel coordinates.
type Seam struct {
X int
Y int
}
// NewCarver returns an initialized Carver structure.
func NewCarver(width, height int) *Carver {
return &Carver{
width,
height,
make([]float64, width*height),
}
}
// Get energy pixel value.
func (c *Carver) get(x, y int) float64 {
px := x + y*c.Width
return c.Points[px]
}
// Set energy pixel value.
func (c *Carver) set(x, y int, px float64) {
idx := x + y*c.Width
c.Points[idx] = px
}
// ComputeSeams compute the minimum energy level based on the following logic:
// - traverse the image from the second row to the last row
// and compute the cumulative minimum energy M for all possible
// connected seams for each entry (i, j).
//
// - the minimum energy level is calculated by summing up the current pixel value
// with the minimum pixel value of the neighboring pixels from the previous row.
func (c *Carver) ComputeSeams(img *image.NRGBA, p *Processor) []float64 {
var srcImg *image.NRGBA
newImg := image.NewNRGBA(image.Rect(0, 0, img.Bounds().Dx(), img.Bounds().Dy()))
draw.Draw(newImg, newImg.Bounds(), img, image.ZP, draw.Src)
// Replace the energy map seam values with the stored pixel values each time we add a new seam.
for _, seam := range usedSeams {
for _, as := range seam.ActiveSeam {
newImg.Set(as.X, as.Y, as.Pix)
}
}
sobel := SobelFilter(Grayscale(newImg), float64(p.SobelThreshold))
if p.FaceDetect {
if len(p.Classifier) == 0 {
log.Fatal("Please provide an xml face classifier!")
}
cascadeFile, err := ioutil.ReadFile(p.Classifier)
if err != nil {
log.Fatalf("Error reading the cascade file: %v", err)
}
tmpImg, err := os.OpenFile(TempImage, os.O_CREATE|os.O_WRONLY, 0755)
if err != nil {
log.Fatalf("Cannot access temporary image file: %v", err)
}
if err := jpeg.Encode(tmpImg, img, &jpeg.Options{Quality: 100}); err != nil {
log.Fatalf("Cannot encode temporary image file: %v", err)
}
src, err := pigo.GetImage(TempImage)
if err != nil {
log.Fatalf("Cannot open the image file: %v", err)
}
pixels := pigo.RgbToGrayscale(src)
cols, rows := src.Bounds().Max.X, src.Bounds().Max.Y
cParams := pigo.CascadeParams{
MinSize: 100,
MaxSize: int(math.Max(float64(cols), float64(rows))),
ShiftFactor: 0.1,
ScaleFactor: 1.1,
}
imgParams := pigo.ImageParams{
Pixels: pixels,
Rows: rows,
Cols: cols,
Dim: cols,
}
pigo := pigo.NewPigo()
// Unpack the binary file. This will return the number of cascade trees,
// the tree depth, the threshold and the prediction from tree's leaf nodes.
classifier, err := pigo.Unpack(cascadeFile)
if err != nil {
log.Fatalf("Error reading the cascade file: %v\n", err)
}
// Run the classifier over the obtained leaf nodes and return the detection results.
// The result contains quadruplets representing the row, column, scale and detection score.
faces := classifier.RunCascade(imgParams, cParams)
// Calculate the intersection over union (IoU) of two clusters.
faces = classifier.ClusterDetections(faces, 0.2)
// Range over all the detected faces and draw a white rectangle mask over each of them.
// We need to trick the sobel detector to consider them as important image parts.
for _, face := range faces {
if face.Q > 5.0 {
rect := image.Rect(
face.Col - face.Scale / 2,
face.Row - face.Scale / 2,
face.Col + face.Scale / 2,
face.Row + face.Scale / 2,
)
draw.Draw(sobel, rect, &image.Uniform{color.RGBA{255, 255, 255, 255}}, image.ZP, draw.Src)
}
}
// Capture CTRL-C signal and remove the generated temporary image.
c := make(chan os.Signal, 2)
signal.Notify(c, os.Interrupt, syscall.SIGTERM)
go func() {
for range c {
RemoveTempImage(TempImage)
os.Exit(1)
}
}()
}
if p.BlurRadius > 0 {
srcImg = StackBlur(sobel, uint32(p.BlurRadius))
} else {
srcImg = sobel
}
for x := 0; x < c.Width; x++ {
for y := 0; y < c.Height; y++ {
r, _, _, a := srcImg.At(x, y).RGBA()
c.set(x, y, float64(r)/float64(a))
}
}
var left, middle, right float64
// Traverse the image from top to bottom and compute the minimum energy level.
// For each pixel in a row we compute the energy of the current pixel
// plus the energy of one of the three possible pixels above it.
for y := 1; y < c.Height; y++ {
for x := 1; x < c.Width-1; x++ {
left = c.get(x-1, y-1)
middle = c.get(x, y-1)
right = c.get(x+1, y-1)
min := math.Min(math.Min(left, middle), right)
// Set the minimum energy level.
c.set(x, y, c.get(x, y)+min)
}
// Special cases: pixels are far left or far right
left := c.get(0, y) + math.Min(c.get(0, y-1), c.get(1, y-1))
c.set(0, y, left)
right := c.get(0, y) + math.Min(c.get(c.Width-1, y-1), c.get(c.Width-2, y-1))
c.set(c.Width-1, y, right)
}
return c.Points
}
// FindLowestEnergySeams find the lowest vertical energy seam.
func (c *Carver) FindLowestEnergySeams() []Seam {
// Find the lowest cost seam from the energy matrix starting from the last row.
var min = math.MaxFloat64
var px int
seams := make([]Seam, 0)
// Find the pixel on the last row with the minimum cumulative energy and use this as the starting pixel
for x := 0; x < c.Width; x++ {
seam := c.get(x, c.Height-1)
if seam < min {
min = seam
px = x
}
}
seams = append(seams, Seam{X: px, Y: c.Height - 1})
var left, middle, right float64
// Walk up in the matrix table check the immediate three top pixel seam level
// and add the one which has the lowest cumulative energy.
for y := c.Height - 2; y >= 0; y-- {
middle = c.get(px, y)
// Leftmost seam, no child to the left
if px == 0 {
right = c.get(px+1, y)
if right < middle {
px++
}
// Rightmost seam, no child to the right
} else if px == c.Width-1 {
left = c.get(px-1, y)
if left < middle {
px--
}
} else {
left = c.get(px-1, y)
right = c.get(px+1, y)
min := math.Min(math.Min(left, middle), right)
if min == left {
px--
} else if min == right {
px++
}
}
seams = append(seams, Seam{X: px, Y: y})
}
return seams
}
// RemoveSeam remove the least important columns based on the stored energy (seams) level.
func (c *Carver) RemoveSeam(img *image.NRGBA, seams []Seam, debug bool) *image.NRGBA {
bounds := img.Bounds()
// Reduce the image width with one pixel on each iteration.
dst := image.NewNRGBA(image.Rect(0, 0, bounds.Dx()-1, bounds.Dy()))
for _, seam := range seams {
y := seam.Y
for x := 0; x < bounds.Max.X; x++ {
if seam.X == x {
if debug {
dst.Set(x-1, y, color.RGBA{255, 0, 0, 255})
}
continue
} else if seam.X < x {
dst.Set(x-1, y, img.At(x, y))
} else {
dst.Set(x, y, img.At(x, y))
}
}
}
return dst
}
// AddSeam add new seam.
func (c *Carver) AddSeam(img *image.NRGBA, seams []Seam, debug bool) *image.NRGBA {
var currentSeam []ActiveSeam
var lr, lg, lb uint32
var rr, rg, rb uint32
var py int
bounds := img.Bounds()
dst := image.NewNRGBA(image.Rect(0, 0, bounds.Dx()+1, bounds.Dy()))
for _, seam := range seams {
y := seam.Y
for x := 0; x < bounds.Max.X; x++ {
if seam.X == x {
if debug == true {
dst.Set(x, y, color.RGBA{255, 0, 0, 255})
continue
}
// Calculate the current seam pixel color by averaging the neighboring pixels color.
if y > 0 {
py = y - 1
} else {
py = y
}
if x > 0 {
lr, lg, lb, _ = img.At(x-1, py).RGBA()
} else {
lr, lg, lb, _ = img.At(x, y).RGBA()
}
if y < bounds.Max.Y-1 {
py = y + 1
} else {
py = y
}
if x < bounds.Max.X-1 {
rr, rg, rb, _ = img.At(x+1, py).RGBA()
} else {
rr, rg, rb, _ = img.At(x, y).RGBA()
}
alr, alg, alb := (lr+rr)/2, (lg+rg)/2, (lb+rb)/2
dst.Set(x, y, color.RGBA{uint8(alr >> 8), uint8(alg >> 8), uint8(alb >> 8), 255})
// Append the current seam position and color to the existing seams.
// To avoid picking the same optimal seam over and over again,
// each time we detect an optimal seam we assign a large positive value
// to the corresponding pixels in the energy map.
// We will increase the seams weight by duplicating the pixel value.
currentSeam = append(currentSeam,
ActiveSeam{Seam{x + 1, y},
color.RGBA{
R: uint8((alr + alr) >> 8),
G: uint8((alg + alg) >> 8),
B: uint8((alb + alb) >> 8),
A: 255,
},
})
} else if seam.X < x {
dst.Set(x, y, img.At(x-1, y))
dst.Set(x+1, y, img.At(x, y))
} else {
dst.Set(x, y, img.At(x, y))
}
}
}
usedSeams = append(usedSeams, UsedSeams{currentSeam})
return dst
}
// RotateImage90 rotate the image by 90 degree counter clockwise.
func (c *Carver) RotateImage90(src *image.NRGBA) *image.NRGBA {
b := src.Bounds()
dst := image.NewNRGBA(image.Rect(0, 0, b.Max.Y, b.Max.X))
for dstY := 0; dstY < b.Max.X; dstY++ {
for dstX := 0; dstX < b.Max.Y; dstX++ {
srcX := b.Max.X - dstY - 1
srcY := dstX
srcOff := srcY*src.Stride + srcX*4
dstOff := dstY*dst.Stride + dstX*4
copy(dst.Pix[dstOff:dstOff+4], src.Pix[srcOff:srcOff+4])
}
}
return dst
}
// RotateImage270 rotate the image by 270 degree counter clockwise.
func (c *Carver) RotateImage270(src *image.NRGBA) *image.NRGBA {
b := src.Bounds()
dst := image.NewNRGBA(image.Rect(0, 0, b.Max.Y, b.Max.X))
for dstY := 0; dstY < b.Max.X; dstY++ {
for dstX := 0; dstX < b.Max.Y; dstX++ {
srcX := dstY
srcY := b.Max.Y - dstX - 1
srcOff := srcY*src.Stride + srcX*4
dstOff := dstY*dst.Stride + dstX*4
copy(dst.Pix[dstOff:dstOff+4], src.Pix[srcOff:srcOff+4])
}
}
return dst
}
// RemoveTempImage removes the temporary image generated during face detection process.
func RemoveTempImage(tmpImage string) {
// Remove temporary image file.
if _, err := os.Stat(tmpImage); err == nil {
os.Remove(tmpImage)
}
}