Separate out binarize into a package, and start adding tests for it

9 files changed, 95 insertions, 8 deletions

diff --git a/binarize/integralimg.go b/binarize/integralimg.go
index c585d60..75e9cce 100644
--- a/binarize/integralimg.go
+++ b/binarize/integralimg.go
@@ -1,4 +1,4 @@
-package main
+package binarize
 
 import (
 	"image"
diff --git a/binarize/sauvola.go b/binarize/sauvola.go
index bc311ad..ca0acf0 100644
--- a/binarize/sauvola.go
+++ b/binarize/sauvola.go
@@ -1,4 +1,4 @@
-package main
+package binarize
 
 import (
 	"image"
diff --git a/binarize/sauvola_test.go b/binarize/sauvola_test.go
new file mode 100644
index 0000000..12a040a
--- /dev/null
+++ b/binarize/sauvola_test.go
@@ -0,0 +1,88 @@
+package binarize
+
+import (
+	"fmt"
+	"image"
+	"image/draw"
+	"image/png"
+	"os"
+	"testing"
+)
+
+func decode(s string) (*image.Gray, error) {
+	f, err := os.Open(s)
+	defer f.Close()
+	if err != nil {
+		return nil, err
+	}
+	img, err := png.Decode(f)
+	if err != nil {
+		return nil, err
+	}
+	b := img.Bounds()
+	gray := image.NewGray(image.Rect(0, 0, b.Dx(), b.Dy()))
+	draw.Draw(gray, b, img, b.Min, draw.Src)
+	return gray, nil
+}
+
+func imgsequal(img1 *image.Gray, img2 *image.Gray) bool {
+	b := img1.Bounds()
+	if ! b.Eq(img2.Bounds())  {
+		return false
+	}
+	for y := b.Min.Y; y < b.Max.Y; y++ {
+		for x := b.Min.X; x < b.Max.X; x++ {
+			r0, g0, b0, a0 := img1.At(x, y).RGBA()
+			r1, g1, b1, a1 := img2.At(x, y).RGBA()
+			if r0 != r1 {
+				return false
+			}
+			if g0 != g1 {
+				return false
+			}
+			if b0 != b1 {
+				return false
+			}
+			if a0 != a1 {
+				return false
+			}
+		}
+	}
+	return true
+}
+
+func TestBinarization(t *testing.T) {
+	// TODO: test non-integral sauvola method too
+	cases := []struct {
+		name string
+		orig string
+		done string
+		ksize float64
+		wsize int
+	}{
+		{"integralsauvola", "testdata/pg1.png", "testdata/pg1_integralsauvola_k0.5_w41.png", 0.5, 41},
+		{"integralsauvola", "testdata/pg1.png", "testdata/pg1_integralsauvola_k0.5_w19.png", 0.5, 19},
+		{"integralsauvola", "testdata/pg1.png", "testdata/pg1_integralsauvola_k0.3_w19.png", 0.3, 19},
+	}
+
+	for _, c := range cases {
+		t.Run(fmt.Sprintf("%s_%0.1f_%d", c.name, c.ksize, c.wsize), func(t *testing.T) {
+			orig, err := decode(c.orig)
+			if err != nil {
+				t.Errorf("Could not open file %s: %v\n", c.orig, err)
+			}
+			done, err := decode(c.done)
+			if err != nil {
+				t.Errorf("Could not open file %s: %v\n", c.done, err)
+			}
+			new := IntegralSauvola(orig, c.ksize, c.wsize)
+			if ! imgsequal(done, new) {
+				t.Errorf("Binarized %s differs to %s\n", c.orig, c.done)
+			}
+		})
+	}
+}
+
+func TestIntegralImg(t *testing.T) {
+	// TODO: compare mean and stddev between integral and basic methods
+}
diff --git a/binarize/testdata/pg1.png b/binarize/testdata/pg1.png
new file mode 100644
index 0000000..2bcc4b1
--- /dev/null
+++ b/binarize/testdata/pg1.png
diff --git a/binarize/testdata/pg1_integralsauvola_k0.3_w19.png b/binarize/testdata/pg1_integralsauvola_k0.3_w19.png
new file mode 100644
index 0000000..bdf5712
--- /dev/null
+++ b/binarize/testdata/pg1_integralsauvola_k0.3_w19.png
diff --git a/binarize/testdata/pg1_integralsauvola_k0.5_w19.png b/binarize/testdata/pg1_integralsauvola_k0.5_w19.png
new file mode 100644
index 0000000..5db2d9a
--- /dev/null
+++ b/binarize/testdata/pg1_integralsauvola_k0.5_w19.png
diff --git a/binarize/testdata/pg1_integralsauvola_k0.5_w41.png b/binarize/testdata/pg1_integralsauvola_k0.5_w41.png
new file mode 100644
index 0000000..050d037
--- /dev/null
+++ b/binarize/testdata/pg1_integralsauvola_k0.5_w41.png
diff --git a/binarize/util.go b/binarize/util.go
index e7cf0f8..e7fcfe4 100644
--- a/binarize/util.go
+++ b/binarize/util.go
@@ -1,4 +1,4 @@
-package main
+package binarize
 
 import (
 	"image"
diff --git a/binarize/main.go b/sauvola/main.go
index ec99c09..c4b9562 100644
--- a/binarize/main.go
+++ b/sauvola/main.go
@@ -1,8 +1,5 @@
 package main
 
-// TODO: could look into other algorithms, see for examples see
-//       the README at https://github.com/brandonmpetty/Doxa
-
 import (
 	"flag"
 	"fmt"
@@ -12,6 +9,8 @@ import (
 	"image/png"
 	"log"
 	"os"
+
+	"rescribe.xyz/go.git/binarize"
 )
 
 // TODO: do more testing to see how good this assumption is
@@ -21,7 +20,7 @@ func autowsize(bounds image.Rectangle) int {
 
 func main() {
 	flag.Usage = func() {
-		fmt.Fprintf(os.Stderr, "Usage: binarize [-w num] [-k num] inimg outimg\n")
+		fmt.Fprintf(os.Stderr, "Usage: sauvola [-w num] [-k num] inimg outimg\n")
 		flag.PrintDefaults()
 	}
 	wsize := flag.Int("w", 0, "Window size for sauvola algorithm. Set automatically based on resolution if not set.")
@@ -56,7 +55,7 @@ func main() {
 
 	// TODO: come up with a way to set a good ksize automatically
 
-	thresh := IntegralSauvola(gray, *ksize, *wsize)
+	thresh := binarize.IntegralSauvola(gray, *ksize, *wsize)
 
 	f, err = os.Create(flag.Arg(1))
 	if err != nil {