def cut(self, need_save=True):
carbration_x = 0
carbration_y = 0
carbration_width = 0
carbration_height = 0
p = ImageUtil.ImageBinaryFunc(self.file_path)
p.get_pic_calibration_data(p.read_pic_data())
p.get_pic_calibration_value_data()
if p.pic_calibration_value is None:
return
elif self.img.width <= 200*2:
return
else:
carbration_x = p.pic_calibration_value[0]
carbration_y = p.pic_calibration_value[1]
carbration_width = p.pic_calibration_value[2]
carbration_height = p.pic_calibration_value[3]
if carbration_width == 0 or carbration_height == 0:
return
top = 0
bottom = 0
con_region_width = 200*2
kind = self.train_info['kind'][1]
if kind in ['C', 'P', 'W', 'B']:
con_region_width = 300*2
con_region_height = 0
top = carbration_y - 100*2
if top < 0:
top = 0
bottom = carbration_y + carbration_height + 100*2
if bottom > self.img.height - 1:
bottom = self.img.height - 1
con_region_height = 960*2 if bottom - top > 960*2 else bottom - top
if con_region_height <= 0:
return
left = 0
if left + con_region_width > self.img.width:
left = self.img.width - con_region_width
rect = (left, top, left + con_region_width, top + con_region_height)
self.regions.append(rect)
left = self.img.width - con_region_width
if left < 0:
left = 0
if left + con_region_width > self.img.width:
left = self.img.width - con_region_width
rect = (left, top, left + con_region_width, top + con_region_height)
self.regions.append(rect)
if need_save:
self.save()
def cut(self, need_save=True):
axel_y = 0
axel_x_offset = 0
rail_y = 0
carbration_x = 0
carbration_y = 0
carbration_width = 0
carbration_height = 0
p = ImageUtil.ImageBinaryFunc(self.file_path)
data = p.read_pic_data()
p.get_pic_calibration_data(data)
p.get_pic_calibration_value_data()
wheel = p.get_pic_wheel_data(data)
if p.pic_calibration_value is None:
return
elif wheel is None:
return
else:
axel_y = p.pic_calibration_value[4]
axel_x_offset = p.pic_calibration_value[5]
rail_y = p.pic_calibration_value[6]
carbration_x = p.pic_calibration_value[0]
carbration_y = p.pic_calibration_value[1]
carbration_width = p.pic_calibration_value[2] - carbration_x + 1
carbration_height = p.pic_calibration_value[3] - carbration_y + 1
region_x = 0
region_y = 0
region_width = self.img.width
region_height = 288 * 2
if region_height > self.img.height:
return
axel_height = abs(rail_y - axel_y)
region_y = axel_y - axel_height*1.3 - region_height
if region_y < 0:
region_y = 0
if region_y + region_height > self.img.height:
region_y = self.img.height - region_height
rect = (region_x, region_y, region_width, region_height)
region_img = self.img.crop(rect)
last_x = 0
while last_x < region_img.width - 200*2:
if last_x + 1096*2 > region_img.width:
last_x = region_img.width - 1096*2
_rect = (last_x, 0, last_x + 1096*2, region_img.height)
last_x += 1096*2 - 200*2
self.regions.append(_rect)
if need_save:
self.save()
def get_calibration(self):
try:
ibf = ImageUtil.ImageBinaryFunc(self._file)
if ibf is None:
return
pic_data = ibf.read_pic_data()
# 获取图像标定信息
ibf.get_pic_calibration_data(pic_data)
ibf.get_pic_calibration_value_data()
self.calibration_info = ibf.pic_calibration_value
# 获取图像车轴信息
self.wheel_info = ibf.get_pic_wheel_data(pic_data)
except Exception as e:
Util.LOG.error(repr(e))
Util.LOG.debug(self._file)
def cut(self, need_save=True):
carbration_y = 0
carbration_height = 0
p = ImageUtil.ImageBinaryFunc(self.file_path)
p.get_pic_calibration_data(p.read_pic_data())
p.get_pic_calibration_value_data()
if p.pic_calibration_value is None:
return
else:
carbration_y = p.pic_calibration_value[1]
carbration_height = p.pic_calibration_value[3] - carbration_y + 1
rect = (0, carbration_y, self.img.width, carbration_y + carbration_height)
self.regions.append(rect)
if need_save:
self.save()
def cut(self, need_save=True):
axel_y = 0
axel_x_offset = 0
rail_y = 0
p = ImageUtil.ImageBinaryFunc(self.file_path)
data = p.read_pic_data()
p.get_pic_calibration_data(data)
p.get_pic_calibration_value_data()
wheel = p.get_pic_wheel_data(data)
if p.pic_calibration_value is None:
return
elif wheel is None:
return
else:
axel_y = p.pic_calibration_value[4]
axel_x_offset = p.pic_calibration_value[5]
rail_y = p.pic_calibration_value[6]
wheel = [self.img.width - x - axel_x_offset for x in wheel]
wheel.reverse()
if len(wheel) != 4:
return
axel_length = 0
axel_height = abs(rail_y - axel_y)
base_point_x = 0
base_point_y = axel_y
chainpipe_x_ratio = 0
chainpipe_y_ratio = 0.75
chainpipe_width_ratio = 0
chainpipe_height_ratio = 1.35
if self.train_info['side'] == 'ZL':
base_point_x = wheel[3]
axel_length = abs(wheel[3] - wheel[2])
if base_point_x < 0 or base_point_x >= self.img.width:
return
chainpipe_x_ratio = 0.22
chainpipe_width_ratio = 1.25
if self.train_info['kind'][1:3] == 'JSQ':
chainpipe_x_ratio = 0.22
chainpipe_width_ratio = 1.35
elif self.train_info['kind'][1] == 'B':
chainpipe_x_ratio = 0.22
chainpipe_width_ratio = 0.95
elif self.train_info['side'] == 'ZR':
base_point_x = wheel[0]
axel_length = abs(wheel[1] - wheel[0])
if base_point_x < 0 or base_point_x >= self.img.width:
return
chainpipe_x_ratio = -1.48
chainpipe_width_ratio = 1.25
if self.train_info['kind'][1:3] == 'JSQ':
chainpipe_x_ratio = -1.58
chainpipe_width_ratio = 1.35
elif self.train_info['kind'][1] == 'B':
chainpipe_x_ratio = -1.18
chainpipe_width_ratio = 0.95
else:
return # 非走行部图片退出
_x = base_point_x + axel_length * chainpipe_x_ratio
_w = axel_length * chainpipe_width_ratio
_w = 1440 if _w > 1440 else _w
_x = 0 if _x < 0 else _x
_x = self.img.width - 2 if _x > self.img.width -1 else _x
if _x + _w >= self.img.width:
_w = self.img.width - _x
_y = base_point_y - axel_height * chainpipe_y_ratio
_h = axel_height * chainpipe_height_ratio
_h = 720 if _h > 720 else _h
_y = 0 if _y < 0 else _y
_y = self.img.height - 2 if _y > self.img.height -1 else _y
if _y + _h >= self.img.height:
_h = self.img.height - _y
rect = (round(_x), round(_y), round(_x + _w), round(_y + _h))
self.regions.append(rect)
if need_save:
self.save()
def cut(self, need_save=True):
img_top = None
img_bottom = None
carbration_x = 0
carbration_y = 0
carbration_width = 0
carbration_height = 0
p = ImageUtil.ImageBinaryFunc(self.file_path)
p.get_pic_calibration_data(p.read_pic_data())
p.get_pic_calibration_value_data()
if p.pic_calibration_value is None:
return
else:
carbration_x = p.pic_calibration_value[0]
carbration_y = p.pic_calibration_value[1]
carbration_width = p.pic_calibration_value[2] - carbration_x + 1
carbration_height = p.pic_calibration_value[3] - carbration_y + 1
if carbration_width == 0 or carbration_height == 0:
return
region_x = 0
region_y = 0
region_width = self.img.width
region_height = 256*2
if region_height > self.img.height:
return
kind = self.train_info['kind'][1]
# 车顶
y_dis = 20*2
if kind in ['C', 'P']:
region_y = carbration_y - region_height + y_dis
if region_y < 0:
region_y = 0
if region_y + region_height > self.img.height:
region_y = self.img.height - region_height
rect = (region_x, region_y, region_x + region_width, region_y + region_height)
img_top = self.img.crop(rect)
#车底
y_dis = 100*2
if kind in ['G']:
y_dis = 50*2
region_y = carbration_y + carbration_height - y_dis
if region_y < 0:
region_y = 0
if region_y + region_height > self.img.height:
region_y = self.img.height - region_height
rect = (region_x, region_y, region_x + region_width, region_y + region_height)
img_bottom = self.img.crop(rect)
for img in [img_top, img_bottom]:
if img is not None:
last_x = 0
while last_x < img.width - 200*2:
if last_x + 1000*2 > img.width:
last_x = img.width - 1000*2
_rect = (last_x, 0, last_x + 1000*2, img.height)
_new = img.crop(_rect)
last_x += 800*2
self.regions.append(_rect)
if need_save:
self.save(img)
