#!/usr/bin/env python
import tesseract
import numpy as np
import cv2
import cv2.cv as cv
import time
import datetime
from time import strftime
from collections import Counter
import ConfigParser
def ConfigSectionMap(section):
global Config
dict1 = {}
options = Config.options(section)
for option in options:
try:
dict1[option] = Config.get(section, option)
if dict1[option] == -1:
DebugPrint("skip: %s" % option)
except:
print("exception on %s!" % option)
dict1[option] = None
return dict1
def SaveImage(event):
global display
params = list()
params.append(cv.CV_IMWRITE_PNG_COMPRESSION)
params.append(8)
filename = datetime.datetime.now().strftime('%y%m%d%H%M%S_%f') + ".png"
cv2.imwrite(filename, display, params)
def OnClose(event):
global stopOpenCv
stopOpenCv = True
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 getthresholdedimg(hsv):
yellow = cv2.inRange(hsv, np.array((20, 100, 100)), np.array((30, 255, 255)))
blue = cv2.inRange(hsv, np.array((100, 100, 100)), np.array((120, 255, 255)))
both = cv2.add(yellow, blue)
return both
if __name__ == '__main__':
Config = ConfigParser.ConfigParser()
Config.read("./config.ini")
############################################################################
# Pre process params
thresh = ConfigSectionMap("PREPROCESS")['threshold']
erosion_iters = ConfigSectionMap("PREPROCESS")['erode']
most_common_filter = ConfigSectionMap("POSPROCESS")['filter']
drawing = False # true if mouse is pressed
start_x, start_y = -1, -1
end_x, end_y = 1, 1
# flag to stop opencv
stopOpenCv = False
# user assist vars
expected_value = ""
expected_value_desv = 20
############################################################################
# Video capture
cap = cv2.VideoCapture(0)
# Tesseract config
api = tesseract.TessBaseAPI()
api.Init(".", ConfigSectionMap("OCR")['fonttype'], tesseract.OEM_DEFAULT)
api.SetVariable("tessedit_char_whitelist", ConfigSectionMap("OCR")['whitelist'])
api.SetPageSegMode(tesseract.PSM_AUTO)
api.SetVariable("debug_file", "/dev/null")
def nothing(x):
pass
# mouse callback function
def draw_rectangle(event, x, y, flags, param):
global start_x, start_y, end_x, end_y, drawing, expected_value
if event == cv2.EVENT_LBUTTONDOWN:
# menu position
if y < 40:
# menu map
if x > 8 and x < 148:
SaveImage(event)
if x > 153 and x < 190:
OnClose(event)
if x > 195 and x < 252:
print "OpenSource Development: https://github.com/arturaugusto/display_ocr.\nBased on examples availables at https://code.google.com/p/python-tesseract/.\nGPLv2 License"
else:
drawing = True
start_x, start_y = x, y
end_x, end_y = x, y
elif event == cv2.EVENT_LBUTTONUP:
drawing = False
#start_x,start_y = -1,-1
#end_x,end_y = -1,-1
elif event == cv2.EVENT_MOUSEMOVE and drawing:
if y < 40:
end_x, end_y = x, 41
else:
end_x, end_y = x, y
cv2.namedWindow('frame')
# Show black image as selected at first
cv2.namedWindow('selection')
display = np.zeros((200, 200, 3), np.uint8)
cv2.imshow('frame', display)
# GUI
cv2.createTrackbar('Threshold', 'frame', int(thresh), 255, nothing)
cv2.createTrackbar('Erode', 'frame', int(erosion_iters), 4, nothing)
cv2.createTrackbar('Filter', 'frame', int(most_common_filter), 10, nothing)
# menu image
menu = cv2.imread("menu.png")
cv2.setMouseCallback('frame', draw_rectangle)
# stack
meas_stack = []
phase_stack = []
time_stack = []
phase_stack_size = 4
control_moving = False
moviment_trigger = 40
while True:
# Capture frame-by-frame
ret, frame = cap.read()
if frame is None:
break
# add menu
frame_h, frame_w = frame.shape[:2]
frame[0:menu.shape[0], 0:menu.shape[1]] = menu
########################################################################
# ROI
########################################################################
cv2.rectangle(frame, (start_x, start_y), (end_x, end_y), (0, 255, 0), 1)
# selection region
min_x = min(start_x, end_x)
max_x = max(start_x, end_x)
min_y = min(start_y, end_y)
max_y = max(start_y, end_y)
display = frame[min_y:max_y, min_x:max_x]
height, width, channel = display.shape
# Show frame
cv2.imshow('frame', frame)
########################################################################
# OCR work
########################################################################
if (height > 10) and (width > 10):
# Display selection on other window
channel = 1
display = cv2.cvtColor(display, cv2.COLOR_BGR2GRAY)
thresh = cv2.getTrackbarPos('Threshold', 'frame')
display = cv2.threshold(display, thresh, 255, cv2.THRESH_BINARY)[1]
kernel = np.ones((5, 5), np.uint8)
erosion_iters = cv2.getTrackbarPos('Erode', 'frame')
display = cv2.erode(display, kernel, iterations=erosion_iters)
# Show selection
cv2.imshow('selection', display)
#image = wx.ImageFromStream(f)
#bitmap = wx.BitmapFromImage(image)
#static_bitmap.SetBitmap(bitmap)
iplimage = cv.CreateImageHeader((width, height), cv.IPL_DEPTH_8U, channel)
cv.SetData(iplimage, display.tostring(), display.dtype.itemsize * channel * (width))
Recognize(iplimage)
c = cv.WaitKey(20)
if c == "q":
break
if stopOpenCv:
break
# When everything done, release the capture
cap.release()
cv2.destroyAllWindows()
