Commit ba4d509

committed

ocr GetComponentImages

1 parent 2f8ef75 commit ba4d509Copy full SHA for ba4d509

File tree

+35

-1

lines changed

+35

-1

lines changed

Lines changed: 29 additions & 0 deletions

Original file line number	Diff line number	Diff line change
`@@ -0,0 +1,29 @@`
	`1`	`+# -- coding: utf-8 --`
	`2`	`+# @Time : 2017年8月21日 14:49`
	`3`	`+# @Author : play4fun`
	`4`	`+# @File : GetComponentImages-example.py.py`
	`5`	`+# @Software: PyCharm`
	`6`	`+`
	`7`	`+"""`
	`8`	`+GetComponentImages-example.py:`
	`9`	`+"""`
	`10`	`+`
	`11`	`+from PIL import Image`
	`12`	`+from tesserocr import PyTessBaseAPI, RIL`
	`13`	`+`
	`14`	`+with PyTessBaseAPI() as api:`
	`15`	`+ # image = Image.open('/usr/src/tesseract/testing/phototest.tif')`
	`16`	`+ image = Image.open('phototest.tif') # 图片有问题`
	`17`	`+ print(image.format, image.info, image.height, image.width)`
	`18`	`+`
	`19`	`+ api.SetImage(image)`
	`20`	`+ boxes = api.GetComponentImages(RIL.TEXTLINE, True)`
	`21`	`+ print('Found {} textline image components.'.format(len(boxes)))`
	`22`	`+ for i, (im, box, _, _) in enumerate(boxes):`
	`23`	`+ # im is a PIL image object`
	`24`	`+ # box is a dict with x, y, w and h keys`
	`25`	`+ api.SetRectangle(box['x'], box['y'], box['w'], box['h'])`
	`26`	`+ ocrResult = api.GetUTF8Text()`
	`27`	`+ conf = api.MeanTextConf()`
	`28`	`+ print(u"Box[{0}]: x={x}, y={y}, w={w}, h={h}, "`
	`29`	`+ "confidence: {1}, text: {2}").format(i, conf, ocrResult, **box)`

100 KB

Binary file not shown.

37.8 KB

Binary file not shown.

Lines changed: 6 additions & 1 deletion

Original file line number	Diff line number	Diff line change
`@@ -10,7 +10,9 @@`
`10`	`10`	`from tesserocr import PyTessBaseAPI, PSM`
`11`	`11`
`12`	`12`	`with PyTessBaseAPI(psm=PSM.AUTO_OSD) as api:`
`13`		`- image = Image.open("/usr/src/tesseract/testing/eurotext.tif")#No such file`
	`13`	`+ # image = Image.open("/usr/src/tesseract/testing/eurotext.tif")#No such file`
	`14`	`+ # image = Image.open("eurotext.tif")`
	`15`	`+ image = Image.open('phototest.tif')`
`14`	`16`	`api.SetImage(image)`
`15`	`17`	`api.Recognize()`
`16`	`18`
`@@ -20,3 +22,6 @@`
`20`	`22`	`print("WritingDirection: {:d}".format(direction))`
`21`	`23`	`print("TextlineOrder: {:d}".format(order))`
`22`	`24`	`print("Deskew angle: {:.4f}".format(deskew_angle))`
	`25`	`+ #`
	`26`	`+ ocrResult = api.GetUTF8Text()`
	`27`	`+ print('result:\n',ocrResult)`

Comments

(0)