summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorNick White <git@njw.name>2019-10-23 10:29:29 +0100
committerNick White <git@njw.name>2019-10-23 10:29:29 +0100
commit8b5c87c6d4e9ed8e220a4f8732c0cd48e92c7a09 (patch)
tree769c2ba323f85558bb6b1ebe9f607daa34231933
parent69aae6b93dcadd9e4895f86fe661ee80e79dcf9e (diff)
Make bucket-lines and related packages more robust
bucket-lines would crash for any line that didn't have a corresponding image. Lines which weren't grayscale would also cause crashes; now they are just converted to grayscale if necessary. As a bonus, lines in jpeg can also be decoded successfull.
-rw-r--r--bucket-lines/main.go4
-rw-r--r--pkg/hocr/lines.go17
2 files changed, 15 insertions, 6 deletions
diff --git a/bucket-lines/main.go b/bucket-lines/main.go
index 6ae506a..af81b44 100644
--- a/bucket-lines/main.go
+++ b/bucket-lines/main.go
@@ -74,7 +74,9 @@ func main() {
}
for _, l := range newlines {
- lines = append(lines, l)
+ if l.Img != nil {
+ lines = append(lines, l)
+ }
}
}
diff --git a/pkg/hocr/lines.go b/pkg/hocr/lines.go
index e90b0a8..d2983be 100644
--- a/pkg/hocr/lines.go
+++ b/pkg/hocr/lines.go
@@ -5,7 +5,9 @@ package hocr
import (
"image"
- "image/png"
+ "image/draw"
+ _ "image/jpeg"
+ _ "image/png"
"io/ioutil"
"log"
"os"
@@ -48,7 +50,7 @@ func getLineText(l OcrLine) (string) {
return linetext
}
-func parseLineDetails(h Hocr, i image.Image, name string) (line.Details, error) {
+func parseLineDetails(h Hocr, i *image.Gray, name string) (line.Details, error) {
lines := make(line.Details, 0)
for _, l := range h.Lines {
@@ -75,7 +77,7 @@ func parseLineDetails(h Hocr, i image.Image, name string) (line.Details, error)
ln.OcrName = name
if i != nil {
var imgd line.ImgDirect
- imgd.Img = i.(*image.Gray).SubImage(image.Rect(coords[0], coords[1], coords[2], coords[3]))
+ imgd.Img = i.SubImage(image.Rect(coords[0], coords[1], coords[2], coords[3]))
ln.Img = imgd
}
lines = append(lines, ln)
@@ -97,20 +99,25 @@ func GetLineDetails(hocrfn string) (line.Details, error) {
}
var img image.Image
+ var gray *image.Gray
pngfn := strings.Replace(hocrfn, ".hocr", ".png", 1)
pngf, err := os.Open(pngfn)
if err != nil {
log.Println("Warning: can't open image %s\n", pngfn)
} else {
defer pngf.Close()
- img, err = png.Decode(pngf)
+ img, _, err = image.Decode(pngf)
if err != nil {
log.Println("Warning: can't load image %s\n", pngfn)
+ } else {
+ b := img.Bounds()
+ gray = image.NewGray(image.Rect(0, 0, b.Dx(), b.Dy()))
+ draw.Draw(gray, b, img, b.Min, draw.Src)
}
}
n := strings.Replace(filepath.Base(hocrfn), ".hocr", "", 1)
- return parseLineDetails(h, img, n)
+ return parseLineDetails(h, gray, n)
}
func GetLineBasics(hocrfn string) (line.Details, error) {