def leerPlaca(self, binarizado):
"""
Args:
img: imagen que contiene el texto, obtenida con la
funcion cv2.imread.
Returns:
El string del texto en la imagen.
"""
cvmat_image = cv.fromarray(binarizado)
imgbincv = cv.GetImage(cvmat_image)
# GetImage fuente de un error, revisar
# Configuración del tesseract
api = tesseract.TessBaseAPI()
api.Init(".", "eng", tesseract.OEM_DEFAULT)
api.SetVariable('tessedit_char_whitelist',
'0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ')
# PSM_SINGLE_CHAR para lectura caracter a caracter
api.SetPageSegMode(tesseract.PSM_SINGLE_CHAR)
# Enviando imagen binarizada al tesseract
tesseract.SetCvImage(imgbincv, api)
text = api.GetUTF8Text()
text = re.sub(r'\W+', '', text)
return text
def ocr():
import cv2.cv as cv
api = tesseract.TessBaseAPI()
api.Init(".", "eng", tesseract.OEM_DEFAULT)
api.SetPageSegMode(tesseract.PSM_AUTO)
image = cv.LoadImage("eurotext.jpg", cv.CV_LOAD_IMAGE_GRAYSCALE)
tesseract.SetCvImage(image, api)
text = api.GetUTF8Text()
conf = api.MeanTextConf()
print(text, len(text))
print("Cofidence Level: %d %%" % conf)
print("Confidences of All words")
header("Method 1", "*" * 10)
confOfText = api.AllWordConfidences()
print(confOfText)
print("Number of Words:")
print("counted by tesseract: %d" % len(confOfText))
print("counted by me: %d[%d]" % (countWords(text), countWords2(text)))
if len(confOfText) != countWords(text):
print("Why the words counted by tesseract are different from mine!!!!")
header("Method 2", "*" * 10)
confs = tesseract.AllWordConfidences(api)
print(confs, len(confs))
def ocr(self):
if self.last_image is not None:
image = self.last_image
# Add border to keep the characters off the edges
offset = 20
height, width, channel = image.shape
image = cv2.copyMakeBorder(image,
offset,
offset,
offset,
offset,
cv2.BORDER_CONSTANT,
value=(255, 255, 255))
# Init and configure tesseract api
api = tesseract.TessBaseAPI()
api.Init(".", "eng", tesseract.OEM_DEFAULT)
api.SetPageSegMode(tesseract.PSM_AUTO)
api.SetVariable("tessedit_char_whitelist", self.white_list)
# Convert to cv image to to pass to tess api
# Derived from example code here: http://blog.mimvp.com/2015/11/python-ocr-recognition/
height, width, channel = image.shape
iplimage = cv.CreateImageHeader((width, height), cv.IPL_DEPTH_8U,
channel)
cv.SetData(iplimage, image.tostring(),
image.dtype.itemsize * channel * (width))
tesseract.SetCvImage(iplimage, api)
api.Recognize(None)
ri = api.GetIterator()
level = tesseract.RIL_WORD
if (ri):
word = ri.GetUTF8Text(level)
return word
def do_ocr(self, cv2_img):
h, w, c = cv2_img.shape
cv_img = cv.CreateImageHeader((w, h), cv.IPL_DEPTH_8U, c)
cv.SetData(cv_img, cv2_img.tostring(), cv2_img.dtype.itemsize * c * w)
#
tesseract.SetCvImage(cv_img, self.ocr_api)
text = self.ocr_api.GetUTF8Text()
#
return text
def tesseract_cv_ocr(imgname):
image = cv.LoadImage(imgname)
api = tesseract.TessBaseAPI()
api.Init(".", "eng", tesseract.OEM_DEFAULT)
#api.SetPageSegMode(tesseract.PSM_SINGLE_WORD)
api.SetPageSegMode(tesseract.PSM_AUTO)
tesseract.SetCvImage(image, api)
text = api.GetUTF8Text()
conf = api.MeanTextConf()
return text
def ocr_number(img):
api = tesseract.TessBaseAPI()
api.Init("C:\Program Files (x86)\Tesseract-OCR", "eng",
tesseract.OEM_DEFAULT)
api.SetVariable("tessedit_char_whitelist", "0123456789+-")
api.SetPageSegMode(tesseract.PSM_SINGLE_LINE)
cv_im = convert_to_IplImage(img)
tesseract.SetCvImage(cv_im, api)
return api.GetUTF8Text().split("\n")[0] # OCR結果から余計な改行を取り除く
def auto_search():
# click search button
genymotion_session.console.mouse.put_mouse_event_absolute(660,290,0,0,0)
genymotion_session.console.mouse.put_mouse_event_absolute(660,290,0,0,1)
genymotion_session.console.mouse.put_mouse_event_absolute(660,290,0,0,0)
sleep(10)
# processing
subprocess.call("adb shell screencap -p /sdcard/screen.png", shell=True)
subprocess.call("adb pull /sdcard/screen.png /tmp/screen.png", shell=True)
im = Image.open("/tmp/screen.png")
#box = (60, 80, 165, 180)
#box = (53, 72, 140, 165)
box = (57, 75, 140, 138)
loot = im.crop(box).convert('L')
loot = ImageOps.invert(loot)
loot.save("/tmp/loot.png", "png")
api = tesseract.TessBaseAPI()
api.Init("/home/mrtux/app/bin/", "coc",tesseract.OEM_DEFAULT)
api.SetVariable("tessedit_char_whitelist", "0123456789")
api.SetPageSegMode(tesseract.PSM_AUTO)
image = cv.LoadImage("/tmp/loot.png", cv.CV_LOAD_IMAGE_UNCHANGED)
tesseract.SetCvImage(image,api)
text = api.GetUTF8Text()
conf = api.MeanTextConf()
total_loot = text.splitlines()
gold_loot, elixir_loot = total_loot[0:2]
gold_loot_text_element = gold_loot.split(" ")
elixir_loot_text_element = elixir_loot.split(" ")
for i in range(len(gold_loot_text_element)):
if len(gold_loot_text_element[i]) > 3:
gold_loot_text_element[i] = gold_loot_text_element[i][1:]
for i in range(len(elixir_loot_text_element)):
if len(elixir_loot_text_element[i]) > 3:
elixir_loot_text_element[i] = elixir_loot_text_element[i][1:]
gold_expr = gold_loot.find(" ") == 3 and int(gold_loot_text_element[0]) >= 200
elixir_expr = elixir_loot.find(" ") == 3 and int(elixir_loot_text_element[0]) >= 200
print gold_loot
print gold_loot_text_element
print elixir_loot
print elixir_loot_text_element
if gold_expr and elixir_expr:
subprocess.call("mplayer /home/mrtux/app/bin/gun.mp3", shell=True)
api.End()
return True
return False
def __init__(self):
#Inquire Credentials
self.buUn = urllib.quote_plus(raw_input("BU TA Username: "******"BU TA Password: "******"Camera", img)
# OCR
tesseract.SetCvImage(img, self.api)
match = re.search(r"(U\d{8})", self.api.GetUTF8Text())
if match:
BUID = match.groups()[0]
print
# Lookup BUID
profile = self.lookupBUID(BUID)
# Prompt for approval
if profile != False and raw_input("Approve (y/n): ") == "y":
self.approveBUID(BUID)
# If Esc is pressed, quit (Also the fps)
if cv.WaitKey(100) == 27: break
def recognize(image):
api = tesseract.TessBaseAPI()
api.Init(".", "eng", tesseract.OEM_DEFAULT)
api.SetPageSegMode(tesseract.PSM_SINGLE_BLOCK)
tesseract.SetCvImage(image, api)
text = api.GetUTF8Text()
conf = api.MeanTextConf()
image = None
print text
print "Confidence: " + str(conf)
return text, conf
def ocr_singledigit(image):
API.Init(".", "eng", tesseract.OEM_DEFAULT)
API.SetVariable("tessedit_char_whitelist", "123456789")
API.SetPageSegMode(6)
tesseract.SetCvImage(image, API)
CHAR = API.GetUTF8Text()
CHAR = CHAR.replace(" ", "").strip()
if len(CHAR) == 0:
return "x"
return int(CHAR)
def getCaptcha():
grayScale = cv.LoadImage(
'https://my.ebharatgas.com/bharatgas/MyCaptcha.png',
cv.CV_LOAD_IMAGE_GRAYSCALE)
cv.Threshold(gray, gray, 231, 255, cv.CV_THRESH_BINARY)
api = tesseract.TessBaseAPI()
api.Init(".", "eng", tesseract.OEM_DEFAULT)
api.SetVariable("tessedit_char_whitelist",
"0123456789abcdefghijklmnopqrstuvwxyz")
api.SetPageSegMode(tesseract.PSM_SINGLE_WORD)
tesseract.SetCvImage(gray, api)
print(api.GetUTF8Text())
def ocr(self, image):
"""Return OCR result."""
api = tesseract.TessBaseAPI()
api.Init(".", "big", tesseract.OEM_DEFAULT)
api.SetPageSegMode(tesseract.PSM_SINGLE_BLOCK)
h, w = image.shape
w_step = w * image.dtype.itemsize
iplimage = cv.CreateImageHeader((w, h), cv.IPL_DEPTH_8U, 1)
cv.SetData(iplimage, image.tostring(), image.dtype.itemsize * (w))
tesseract.SetCvImage(iplimage, api)
hocr = api.GetHOCRText(0)
return hocr
def OCR2(croplist): OCRlist = [] #print OCRlist for i in range(len(croplist)): print len(croplist) print croplist[i] image1=cv.LoadImage(croplist[i]) tesseract.SetCvImage(image1,api) OCR=api.GetUTF8Text() #conf=api.MeanTextConf() print OCR OCRlist.append(OCR) return OCRlist
def testy(image_path):
## Indent region below to test as function
image=cv.LoadImage(image_path, cv.CV_LOAD_IMAGE_GRAYSCALE)
tesseract.SetCvImage(image,api)
text=api.GetUTF8Text()
word_list=text.split()
#print word_list
##Uncomment the below lines to test function
print "Program gets here"
return word_list
def updateOCR(frame):
api = tesseract.TessBaseAPI()
api.SetVariable("tessedit_char_whitelist", WHITELIST)
api.Init("C:\Program Files (x86)\Tesseract-OCR", "eng", tesseract.OEM_DEFAULT)
api.SetPageSegMode(tesseract.PSM_AUTO)
bitmap = cv2.cv.CreateImageHeader((frame.shape[1], frame.shape[0]), cv2.IPL_DEPTH_8U, 3)
cv2.cv.SetData(bitmap, frame.tostring(), frame.dtype.itemsize * 3 * frame.shape[1])
tesseract.SetCvImage(bitmap, api)
text = api.GetUTF8Text()
confidence = api.MeanTextConf()
charMatrix = api.GetBoxText(0).split('\n')
return text, confidence, charMatrix
def ocr(self, image, area, factor):
api = tesseract.TessBaseAPI()
api.Init(".", "big", tesseract.OEM_DEFAULT)
api.SetPageSegMode(tesseract.PSM_SINGLE_BLOCK)
h, w = image.shape
w_step = w * image.dtype.itemsize
iplimage = cv.CreateImageHeader((w, h), cv.IPL_DEPTH_8U, 1)
cv.SetData(iplimage, image.tostring(), image.dtype.itemsize * (w))
tesseract.SetCvImage(iplimage, api)
api.SetRectangle(int(area[0][0] * factor), int(area[0][1] * factor),
int(area[1][0] * factor) - int(area[0][0] * factor),
int(area[1][1] * factor) - int(area[0][1] * factor))
res = self.hocrToObject(api.GetHOCRText(0), area, factor)
return res
def convert_file(filename):
#path = str(app.config['UPLOAD_FOLDER']+filename)
path = str('/home/engineer/htdocs/stop/webapi/uploads/' + filename).lower()
try:
image = cv.LoadImage(path, cv.CV_LOAD_IMAGE_GRAYSCALE)
except Exception as e:
return str("Error ") + str(e)
api = tesseract.TessBaseAPI()
api.Init(".", "eng", tesseract.OEM_DEFAULT)
api.SetPageSegMode(tesseract.PSM_AUTO)
tesseract.SetCvImage(image, api)
text = api.GetUTF8Text()
conf = api.MeanTextConf()
return jsonify({'output': str(text)})
def img2text_method2(image_path):
print "Using method 2 to generate text from image..."
api = tesseract.TessBaseAPI()
api.SetOutputName("outputName")
api.Init(".", "eng", tesseract.OEM_DEFAULT)
api.SetPageSegMode(tesseract.PSM_AUTO)
image = cv.LoadImage(image_path, cv.CV_LOAD_IMAGE_GRAYSCALE)
tesseract.SetCvImage(image, api)
text = api.GetUTF8Text()
word_list = text.split()
print word_list
return word_list
def ocr_udemae(img):
img = binarize(img)
img = erode(img)
# cv2.imshow('ocr_udemae', img)
# cv2.waitKey(500)
api = tesseract.TessBaseAPI()
api.Init("C:\Program Files (x86)\Tesseract-OCR", "eng",
tesseract.OEM_DEFAULT)
api.SetVariable("tessedit_char_whitelist", "SABC+-")
api.SetPageSegMode(tesseract.PSM_SINGLE_LINE)
cv_im = convert_to_IplImage(img)
tesseract.SetCvImage(cv_im, api)
return api.GetUTF8Text().split("\n")[0] # OCR結果から余計な改行を取り除く
def img2txt(image_path):
#print "entered img2txt"
#print "CONVERTING IMAGE TO TEXT"
#print "---------------------------------------------------"
image=cv.LoadImage(image_path, cv.CV_LOAD_IMAGE_GRAYSCALE)
#print "set image"
tesseract.SetCvImage(image,api)
#print "did the tesseract thing"
text=api.GetUTF8Text()
#print "actually got the text"
conf=api.MeanTextConf()
## Remove every '\n' PUT IN BY TESS and put everything back together with a space!
page_text=" ".join(text.splitlines())
#print "Text for " + image_path + " : " + text
return page_text
def ocr_img(self, imgname):
"""
To use a non-standard language pack named foo.traineddata, set the TESSDATA_PREFIX environment variable so the file can be found at TESSDATA_PREFIX/tessdata/foo.traineddata and give Tesseract the argument -l foo.
"""
image = cv.LoadImage(imgname, cv.CV_LOAD_IMAGE_GRAYSCALE)
api = tesseract.TessBaseAPI()
api.Init(".", "eng", tesseract.OEM_DEFAULT)
api.SetPageSegMode(tesseract.PSM_AUTO)
tesseract.SetCvImage(image, api)
text = api.GetUTF8Text()
conf = api.MeanTextConf()
return text
def Recognize(iplimage):
global meas_stack
tesseract.SetCvImage(iplimage, api)
try:
full_text = api.GetUTF8Text()
except AttributeError:
full_text = api.GetUNLVText().replace("^", "")
conf = api.MeanTextConf()
# Ger the first line found by tesseract
for index, text in enumerate(full_text.split('\n')):
# Some char filter
text = text.replace(" ", "")
for char in ConfigSectionMap("POSPROCESS")['strip']:
text = text.replace(char, "")
try:
text_val = float(text)
# handle OCRed value if exists an expected value prvided by user
if expected_value != "":
up_limit = (float(expected_value)) * (1 + (float(expected_value_desv) / 100))
dn_limit = (float(expected_value)) * (1 - (float(expected_value_desv) / 100))
if (
len(text) > 0 and
text_val > dn_limit and
text_val < up_limit
):
pass
else:
return 0
# most common filter valur
most_common_filter_pos = cv2.getTrackbarPos('Filter', 'frame')
# add last text
meas_stack.append(text)
if len(meas_stack) > most_common_filter_pos:
# remove old
meas_stack = meas_stack[-(most_common_filter_pos + 1):]
# count most frequent value
count = Counter(meas_stack)
out = count.most_common()[0][0]
## show if the last is the most common
# if out == meas_stack[-1]:
print "Timestamp: " + datetime.datetime.now().strftime('%y%m%d%H%M%S_%f')
print "Line " + str(index)
print out
except:
pass
def ocr_sxx(filename):
#####Image preprocess#####
im = cv2.imread(filename)
#convert to grayscale and apply Gaussian filtering
im_gray = cv2.cvtColor(im, cv2.COLOR_BGR2GRAY)
#cv2.imshow('result1',im_gray)
# Threshold the image
ret, im_th = cv2.threshold(im_gray, 70, 255, cv2.THRESH_BINARY)
#cv2.imshow('result2',im_th)
#remove noise
im_th = cv2.medianBlur(im_th, 3)
#check result
#cv2.imshow('result',im_th)
#save the image in local computer
cv2.imwrite('result.png', im_th)
cv2.waitKey(0) # Waits forever for user to press any key
cv2.destroyAllWindows()
#########----OCR Part----########
image = cv.LoadImage("result.png", cv.CV_LOAD_IMAGE_GRAYSCALE)
directory = "tessdata/"
########choose language############
InputLan = "eng"
api = tesseract.TessBaseAPI()
fileName = InputLan + '.traineddata'
fullPath = directory + fileName
api.Init(".", InputLan, tesseract.OEM_DEFAULT)
#api.SetPageSegMode(tesseract.PSM_SINGLE_WORD)
api.SetPageSegMode(tesseract.PSM_AUTO)
tesseract.SetCvImage(image, api)
text = api.GetUTF8Text()
conf = api.MeanTextConf()
timage = None
print len(text)
#text = text.decode('utf-8') #from str to unicode
print text.decode('utf-8')
#print conf
########delete /n ############
line = text.replace('\n', ' ')
#line = line.decode('utf-8')
print line.decode('utf-8')
def ocr(self, image, area, factor):
api = tesseract.TessBaseAPI()
#print self.lang
if self.lang == "big" or self.lang == "eng":
api.Init(".","big",tesseract.OEM_DEFAULT)
else:
api.Init(".", str(self.lang), tesseract.OEM_DEFAULT)
api.SetPageSegMode(tesseract.PSM_SINGLE_BLOCK)
h,w = image.shape
w_step = w*image.dtype.itemsize
iplimage = cv.CreateImageHeader((w,h), cv.IPL_DEPTH_8U, 1)
cv.SetData(iplimage, image.tostring(),image.dtype.itemsize * (w))
tesseract.SetCvImage(iplimage,api)
ocr_x2 = int((area[1][0]-int((area[1][0]-area[0][0])*0.17))*factor)
api.SetRectangle(int(area[0][0]*factor), int(area[0][1]*factor),
ocr_x2-int(area[0][0]*factor),
int(area[1][1]*factor)-int(area[0][1]*factor))
res = self.hocrToObject(api.GetHOCRText(0).decode('utf-8'), area, factor)
return res
def tesseract_native(fpath, lang, block=True):
"""
Use tesseract Python API to process `fpath`
pre-segmented image
"""
api = tesseract.TessBaseAPI()
api.Init(".", lang, tesseract.OEM_DEFAULT)
if block:
api.SetPageSegMode(tesseract.PSM_SINGLE_BLOCK)
else:
api.SetPageSegMode(tesseract.PSM_AUTO)
for opt, val in tesseract_opts.items():
api.SetVariable(opt, val)
img = cv.LoadImage(fpath, iscolor=False)
tesseract.SetCvImage(img, api)
text = api.GetUTF8Text()
return text
def leerPlaca(self, binarizado):
"""
Args:
img: imagen que contiene el texto, obtenida con la
funcion cv2.imread.
Returns:
El string del texto en la imagen.
"""
cvmat_image = cv.fromarray(binarizado)
imgbincv = cv.GetImage(cvmat_image)
# GetImage fuente de un error, revisar
# Configuración del tesseract
api = tesseract.TessBaseAPI()
api.Init(".", "eng", tesseract.OEM_DEFAULT)
api.SetPageSegMode(tesseract.PSM_AUTO)
# Enviando imagen binarizada al tesseract
tesseract.SetCvImage(imgbincv, api)
text = api.GetUTF8Text()
text = re.sub(r'\W+', '', text)
return text
def getText(image0):
#thicken the border in order to make tesseract feel happy
offset = textOffset
height, width, channel = image0.shape
image1=cv2.copyMakeBorder(image0, offset, offset,offset,offset,\
cv2.BORDER_CONSTANT,value=rgbWhite)
#set up tesseract API
api = tesseract.TessBaseAPI()
api.Init(".", "eng", tesseract.OEM_DEFAULT)
api.SetPageSegMode(tesseract.PSM_AUTO)
height1, width1, channel1 = image1.shape
#star text extraction
cvmat_image = cv.fromarray(image1)
iplimage = cv.GetImage(cvmat_image)
#extract texts
tesseract.SetCvImage(iplimage, api)
text = api.GetUTF8Text()
conf = api.MeanTextConf()
return text, conf
def __get_text_from_image(saved_image_path, coord):
try:
import cv2
import cv2.cv as cv
import tesseract
except:
raise ImportError("tesseract library for python required")
api = tesseract.TessBaseAPI()
api.Init(".", "eng", tesseract.OEM_DEFAULT)
api.SetPageSegMode(tesseract.PSM_AUTO)
image0 = cv2.imread(saved_image_path, cv.CV_LOAD_IMAGE_GRAYSCALE)
if image0 is None:
raise cv2.error("Image for text matching was NoneType")
# x1 = coord[0], y1 = coord[1], x2 = coord[2], y2 = coord[3]
image1 = image0[coord[1]:coord[3], coord[0]:coord[2]]
height1, width1 = image1.shape
iplimage = cv.CreateImageHeader((width1, height1), cv.IPL_DEPTH_8U, 1)
cv.SetData(iplimage, image1.tostring(), image1.dtype.itemsize * (width1))
tesseract.SetCvImage(iplimage, api)
return api.GetUTF8Text()
}
captcha = 'http://158.69.76.135/captcha.php'
c = requests.get(captcha, headers=headers)
with tempfile.NamedTemporaryFile(delete=False) as temp:
temp.write(c.raw.read())
temp.seek(0)
temp.close()
gray = cv.LoadImage(temp.name, cv.CV_LOAD_IMAGE_GRAYSCALE)
cv.Threshold(gray, gray, 231, 255, cv.CV_THRESH_BINARY)
api = tesseract.TessBaseAPI()
api.Init(".", "eng", tesseract.OEM_DEFAULT)
api.SetVariable("tessedit_char_whitelist",
"0123456789abcdefghijklmnopqrstuvwxyz")
api.SetPageSegMode(tesseract.PSM_SINGLE_WORD)
tesseract.SetCvImage(gray, api)
captcha = api.GetUTF8Text()
posturl = "http://158.69.76.135/level3.php"
headers = {
'Referer': 'http://158.69.76.135/level3.php',
'User-Agent': 'Mozilla/4.01 [en] (Win95; I)'
}
get = requests.get(posturl, headers=headers)
for i in range(1024):
get = requests.post(posturl,
data={
'id': '923',
#/usr/bin/env python
import numpy as np
import cv2
import cv2.cv as cv
import tesseract
image = cv2.imread("temp.png")
image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
blurred = cv2.GaussianBlur(image, (5, 5), 0)
(T, tresh) = cv2.threshold(blurred, 155, 255, cv2.THRESH_BINARY)
cv2.imshow("Avg GaussianBlur", np.hstack([blurred, tresh]))
cv2.waitKey(0)
api = tesseract.TessBaseAPI()
api.Init(".", "eng", tesseract.OEM_DEFAULT)
api.SetPageSegMode(tesseract.PSM_SINGLE_BLOCK)
height1, width1 = tresh.shape
channel1 = 1
image = cv.CreateImageHeader((width1, height1), cv.IPL_DEPTH_8U, channel1)
cv.SetData(image, tresh.tostring(), tresh.dtype.itemsize * channel1 * (width1))
tesseract.SetCvImage(image, api)
text = api.GetUTF8Text()
conf = api.MeanTextConf()
image = None
print "..............."
print "Ocred Text: %s" % text
print "Cofidence Level: %d %%" % conf