def makeClusterBack(self, isInterrupt):
'''
brief: make the clustering result
:param isInterrupt: the running elapsed of per frame is exceed or not
:return: back, the clustering mask
area, the clustering area percentage
'''
back, circles = np.zeros((self.h, self.w), dtype=np.uint8), []
for i, rect in enumerate(self.hisrects):
if self.ovlcount[i] < self.retainov:
continue
ismerge, (xi, yi, wi, hi) = False, rect
centeri, radiusi = (int(xi+wi/2), int(yi+hi/2)), int((wi+hi)/4)
for j, (centerj, radiusj) in enumerate(circles):
thresr = 1.25 * (radiusi + radiusj)
if util.distance(centeri, centerj) < thresr:
ismerge = True
cv2.line(back, centeri, centerj, 255, int(thresr))
if not ismerge:
circles.append([centeri, radiusi])
backrs, backsize = cv2.resize(back, (int(self.w/8), int(self.h/8))), self.w * self.h / 64
area = 1.0 if isInterrupt else (len(np.argwhere(backrs > 8)) / backsize)
return (cv2.merge([cv2.Sobel(back, cv2.CV_8UC1, 1, 1)] * 3), area)
def distance_table(self, data_table, cols, d_function):
data_table[cols] = data_table.loc[:, cols].astype('float32')
return pd.DataFrame(scipy.spatial.distance.squareform(
util.distance(data_table.loc[:, cols], d_function)),
columns=data_table.index,
index=data_table.index).astype('float32')
def create_distance_table(data_table: pd.DataFrame, cols: List[str],
d_function: str) -> pd.DataFrame:
"""
Create distance table between rows in the data table. Only cols are considered and the specified distance
function is used to compute the distance .
:param data_table: DataFrame to calculate distance matrix for.
:param cols: Cols to use for calculating distance between rows.
:param d_function: Distance function to use for calculation. By now only euclidean is supported.
:return: NxN matrix with distance from each point to each where N is the number of rows.
"""
data_table[cols] = data_table.loc[:, cols].astype('float32')
print(
'Calculating distance matrix, this may take a while and your computer may be slower.'
)
return pd.DataFrame(scipy.spatial.distance.squareform(
util.distance(data_table.loc[:, cols], d_function)),
columns=data_table.index,
index=data_table.index).astype('float32')
def distance_table(self, data_table, cols, d_function):
return pd.DataFrame(scipy.spatial.distance.squareform(
util.distance(data_table.ix[:, cols], d_function)),
columns=data_table.index,
index=data_table.index)