def test_get_slide_name(self):
db = DBConnection("../data_base/test_base.db")
db.insert_slide("1","1.jpg")
db.insert_object("1","LOL")
disp = InfoDisplayer(db, "TestImages/")
name = disp.get_slide_name(1)
self.assertTrue(name == "1.jpg")
db.clear()
def test_display(self):
db = DBConnection("../data_base/test_base.db")
db.insert_slide("1","1.jpg")
db.insert_object("1","LOL")
disp = InfoDisplayer(db, "TestImages/")
im = cv2.imread("TestImages/1.jpg")
im2 = cv2.imread("TestImages/2.jpg")
im2 = disp.display(1,im2)
db.clear()
self.assertTrue(len(im) == len(im2))
for x in range(im.shape[0]):
for y in range(im.shape[1]):
self.assertTrue(im[x][y][0]==im2[x][y][0])
def __init__(self, db, size=6):
self.info_displayer = InfoDisplayer(db)
self.size = size
self.frame = None
self.contours = None
self.th = None
self.detected = False
self.id = None
def test_get_first(self):
db = DBConnection("../data_base/test_base.db")
db.insert_slide("1","1.jpg")
db.insert_slide("2","2.jpg")
db.insert_object("1","LOL")
db.insert_object("2","LOL")
ob1 = (1,"LOL")
ob2 = (2,"LOL")
ob3 = (3,"LOL")
disp = InfoDisplayer(db, "TestImages/")
out = disp.get_first(ob1)
self.assertTrue(out == "1.jpg")
out = disp.get_first(ob2)
self.assertTrue(out == "2.jpg")
out = disp.get_first(ob3)
self.assertTrue(out == "")
db.clear()
class DataMatrixDetector:
def __init__(self, db, size=6):
self.info_displayer = InfoDisplayer(db)
self.size = size
self.frame = None
self.contours = None
self.th = None
self.detected = False
self.id = None
def set_template(self, template):
"""
@param template string like 'path/filename' of template image
Function set template that will be used to detect matrix.
"""
img = cv2.imread(template, 0)
blur = cv2.GaussianBlur(img, (5, 5), 0)
ret, thtemplate = cv2.threshold(blur, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
im, template_contours, hierarchy2 = cv2.findContours(thtemplate, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
self.contours = [contour for contour in template_contours if 5000 < cv2.contourArea(contour) < 100000]
logging.debug("Template "+template + " set. Contours length: "+ str(len(self.contours)) )
def check_matrix(self, matrix):
"""
@param matrix
Return true if matrix is data matrix or false if not
Function verify matrix are data matrix.
"""
for i in range(self.size):
if (matrix[0][i] + matrix[-1][i] == i % 2 or matrix[0][i] + matrix[-1][i] == (i + 1) % 2) and (
matrix[i][0] + matrix[i][-1] == i % 2 or matrix[i][0] + matrix[i][-1] == (i + 1) % 2):
pass
else:
logging.debug("Matrix detection failed. Matrix passed to function " + str(matrix))
return False
for i in range(self.size):
for j in range(self.size):
if matrix[i][j] > 1:
logging.debug("Matrix detection failed. Matrix passed to function " + str(matrix))
return False
logging.info("Matrix detected : " + str(matrix))
return True
def turn_matrix(self, matrix):
"""
@param matrix
@return well oriented matrix
Function make matrix well oriented.
"""
if matrix[0][0] == 1:
tab = [[(matrix[-j - 1][i] + 1) % 2 for j in range(self.size)] for i in range(self.size)]
elif matrix[0][1] == 1:
tab = [[(matrix[i][j] + 1) % 2 for j in range(self.size)] for i in range(self.size)]
elif matrix[1][0] == 1:
tab = [[(matrix[i][j] + 1) % 2 for j in range(self.size)] for i in range(self.size)]
else:
tab = [[(matrix[j][-1 - i] + 1) % 2 for j in range(self.size)] for i in range(self.size)]
return tab
def read_matrix(self, rect):
"""
@param rect is two dimention array with point of a rectangle surrounding matrix area on image
@return true if matrix is read succesfully or false if not
Function read matrix in selected area.
"""
wx = (rect[1][0] - rect[0][0]) / self.size
wy = (rect[1][1] - rect[0][1]) / self.size
vx = (rect[3][0] - rect[0][0]) / self.size
vy = (rect[3][1] - rect[0][1]) / self.size
first = [rect[0][0] + (wx + vx) / 2, rect[0][1] + (wy + vy) / 2]
try:
matrix = [[self.th[first[1] + i * wy + j * vy][first[0] + i * wx + j * vx] for i in range(self.size)] for j
in range(self.size)]
except IndexError:
return None
if self.check_matrix(matrix):
print(True)
self.detected = True
print(matrix)
self.id = DataMatrixCrypto.decode(self.turn_matrix(matrix))
print(matrix)
print(self.id)
return True
else:
self.detected = False
return False
def take_contours(self):
"""
@return contours of objects find on image
Function find contours on image
"""
gray = cv2.cvtColor(self.frame, cv2.COLOR_BGR2GRAY)
blur = cv2.GaussianBlur(gray, (5, 5), 0)
ret, self.th = cv2.threshold(blur, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
im2, contours, hierarchy = cv2.findContours(self.th, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
return [contour for contour in contours if 1000 < cv2.contourArea(contour) < 150000]
def detect_matrix(self, current_frame):
"""
@param current_frame is image where matrix is detecting
Function detext matrix on image.
"""
self.frame = current_frame[0]
contours = self.take_contours()
for cnt, cnt2 in zip(contours, self.contours):
val = cv2.matchShapes(cnt, cnt2, 1, 0.0)
if val < 0.2:
rect = cv2.minAreaRect(cnt)
box = cv2.boxPoints(rect)
box = np.int0(box)
if self.read_matrix(box):
cv2.drawContours(self.frame, [box], 0, (0, 0, 255), 2)
cv2.drawContours(self.frame, cnt, -1, (255, 0, 0), 3)
if self.detected:
img = self.info_displayer.display(self.id, self.frame)
logging.debug("Data Matrix detected")
current_frame = [self.frame]
