初版
This commit is contained in:
leon
parent
60985d84bc
commit
304d9422e3
1
.gitignore
vendored
1
.gitignore
vendored
@ -1 +1,2 @@
|
||||
.idea/
|
||||
/test.py
|
||||
|
||||
BIN
bim_im.png
BIN
bim_im.png
Binary file not shown.
|
Before Width: | Height: | Size: 69 KiB After Width: | Height: | Size: 115 KiB |
@ -1,202 +0,0 @@
|
||||
[
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.9336103200912476,
|
||||
"x": 325,
|
||||
"y": 217,
|
||||
"width": 68,
|
||||
"height": 71
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.9302101731300354,
|
||||
"x": 561,
|
||||
"y": 222,
|
||||
"width": 56,
|
||||
"height": 66
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.9269363284111023,
|
||||
"x": 440,
|
||||
"y": 115,
|
||||
"width": 54,
|
||||
"height": 55
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.9259827136993408,
|
||||
"x": 495,
|
||||
"y": 269,
|
||||
"width": 41,
|
||||
"height": 42
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.915646493434906,
|
||||
"x": 430,
|
||||
"y": 321,
|
||||
"width": 166,
|
||||
"height": 38
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.9102611541748047,
|
||||
"x": 276,
|
||||
"y": 94,
|
||||
"width": 43,
|
||||
"height": 41
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.9072883129119873,
|
||||
"x": 451,
|
||||
"y": 220,
|
||||
"width": 80,
|
||||
"height": 38
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.9049834609031677,
|
||||
"x": 179,
|
||||
"y": 323,
|
||||
"width": 183,
|
||||
"height": 38
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.8983769416809082,
|
||||
"x": 512,
|
||||
"y": 102,
|
||||
"width": 40,
|
||||
"height": 41
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.8979458212852478,
|
||||
"x": 71,
|
||||
"y": 216,
|
||||
"width": 67,
|
||||
"height": 66
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.8557734489440918,
|
||||
"x": 579,
|
||||
"y": 157,
|
||||
"width": 31,
|
||||
"height": 34
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.8004537224769592,
|
||||
"x": 201,
|
||||
"y": 218,
|
||||
"width": 87,
|
||||
"height": 39
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.7972114682197571,
|
||||
"x": 195,
|
||||
"y": 110,
|
||||
"width": 52,
|
||||
"height": 55
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.7899783849716187,
|
||||
"x": 350,
|
||||
"y": 97,
|
||||
"width": 43,
|
||||
"height": 41
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.789060115814209,
|
||||
"x": 34,
|
||||
"y": 93,
|
||||
"width": 42,
|
||||
"height": 39
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.7720834612846375,
|
||||
"x": 573,
|
||||
"y": 107,
|
||||
"width": 37,
|
||||
"height": 40
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.753940761089325,
|
||||
"x": 247,
|
||||
"y": 267,
|
||||
"width": 46,
|
||||
"height": 45
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.7526902556419373,
|
||||
"x": 304,
|
||||
"y": 170,
|
||||
"width": 36,
|
||||
"height": 34
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.7418428063392639,
|
||||
"x": 103,
|
||||
"y": 142,
|
||||
"width": 37,
|
||||
"height": 35
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.7005344033241272,
|
||||
"x": 53,
|
||||
"y": 167,
|
||||
"width": 36,
|
||||
"height": 36
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.6595599055290222,
|
||||
"x": 353,
|
||||
"y": 146,
|
||||
"width": 38,
|
||||
"height": 37
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.6562734842300415,
|
||||
"x": 103,
|
||||
"y": 91,
|
||||
"width": 45,
|
||||
"height": 41
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.6161856651306152,
|
||||
"x": 540,
|
||||
"y": 175,
|
||||
"width": 33,
|
||||
"height": 36
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.07473669946193695,
|
||||
"x": 122,
|
||||
"y": 33,
|
||||
"width": 314,
|
||||
"height": 358
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.056293122470378876,
|
||||
"x": 0,
|
||||
"y": 216,
|
||||
"width": 43,
|
||||
"height": 37
|
||||
}
|
||||
]
|
||||
130
data_sub/test_2/data_sub.json
Normal file
130
data_sub/test_2/data_sub.json
Normal file
@ -0,0 +1,130 @@
|
||||
[
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.9288522601127625,
|
||||
"x": 212,
|
||||
"y": 112,
|
||||
"width": 54,
|
||||
"height": 52
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.9224857687950134,
|
||||
"x": 337,
|
||||
"y": 215,
|
||||
"width": 64,
|
||||
"height": 65
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.9219771027565002,
|
||||
"x": 289,
|
||||
"y": 96,
|
||||
"width": 40,
|
||||
"height": 40
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.92093825340271,
|
||||
"x": 94,
|
||||
"y": 214,
|
||||
"width": 68,
|
||||
"height": 67
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.8973387479782104,
|
||||
"x": 358,
|
||||
"y": 96,
|
||||
"width": 41,
|
||||
"height": 40
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.8777830600738525,
|
||||
"x": 217,
|
||||
"y": 213,
|
||||
"width": 88,
|
||||
"height": 39
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.8754808902740479,
|
||||
"x": 196,
|
||||
"y": 314,
|
||||
"width": 180,
|
||||
"height": 39
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.8678174614906311,
|
||||
"x": 262,
|
||||
"y": 262,
|
||||
"width": 45,
|
||||
"height": 42
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.7945911288261414,
|
||||
"x": 362,
|
||||
"y": 147,
|
||||
"width": 36,
|
||||
"height": 35
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.7796808481216431,
|
||||
"x": 0,
|
||||
"y": 116,
|
||||
"width": 38,
|
||||
"height": 55
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.7364567518234253,
|
||||
"x": 0,
|
||||
"y": 216,
|
||||
"width": 65,
|
||||
"height": 36
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.5554796457290649,
|
||||
"x": 48,
|
||||
"y": 311,
|
||||
"width": 87,
|
||||
"height": 42
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.5389373898506165,
|
||||
"x": 315,
|
||||
"y": 168,
|
||||
"width": 37,
|
||||
"height": 36
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.38828131556510925,
|
||||
"x": 55,
|
||||
"y": 98,
|
||||
"width": 43,
|
||||
"height": 41
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.3368460536003113,
|
||||
"x": 123,
|
||||
"y": 98,
|
||||
"width": 44,
|
||||
"height": 39
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.07753150165081024,
|
||||
"x": 74,
|
||||
"y": 168,
|
||||
"width": 40,
|
||||
"height": 37
|
||||
}
|
||||
]
|
||||
BIN
data_sub/test_2/wide_image.png
Normal file
BIN
data_sub/test_2/wide_image.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 444 KiB |
90
data_sub/test_3/return_objs_data.json
Normal file
90
data_sub/test_3/return_objs_data.json
Normal file
@ -0,0 +1,90 @@
|
||||
[
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.917050302028656,
|
||||
"x": 189,
|
||||
"y": 415,
|
||||
"width": 196,
|
||||
"height": 40
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.9155016541481018,
|
||||
"x": 214,
|
||||
"y": 212,
|
||||
"width": 53,
|
||||
"height": 48
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.9061460494995117,
|
||||
"x": 338,
|
||||
"y": 309,
|
||||
"width": 68,
|
||||
"height": 71
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.9045385122299194,
|
||||
"x": 92,
|
||||
"y": 309,
|
||||
"width": 70,
|
||||
"height": 69
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.8958920240402222,
|
||||
"x": 262,
|
||||
"y": 358,
|
||||
"width": 46,
|
||||
"height": 45
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.8823522329330444,
|
||||
"x": 219,
|
||||
"y": 311,
|
||||
"width": 85,
|
||||
"height": 37
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.8750882744789124,
|
||||
"x": 289,
|
||||
"y": 197,
|
||||
"width": 39,
|
||||
"height": 36
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.7151791453361511,
|
||||
"x": 0,
|
||||
"y": 212,
|
||||
"width": 48,
|
||||
"height": 48
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.569022536277771,
|
||||
"x": 354,
|
||||
"y": 198,
|
||||
"width": 42,
|
||||
"height": 36
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.4601861834526062,
|
||||
"x": 314,
|
||||
"y": 265,
|
||||
"width": 36,
|
||||
"height": 36
|
||||
},
|
||||
{
|
||||
"label": 0,
|
||||
"prob": 0.3012613356113434,
|
||||
"x": 360,
|
||||
"y": 244,
|
||||
"width": 35,
|
||||
"height": 33
|
||||
}
|
||||
]
|
||||
BIN
data_sub/test_3/wide_image.png
Normal file
BIN
data_sub/test_3/wide_image.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 414 KiB |
138
search.py
138
search.py
@ -1,6 +1,6 @@
|
||||
import cv2
|
||||
|
||||
from utils import filter_points
|
||||
from utils import filter_rectangle
|
||||
|
||||
print(cv2.__version__) # 4.9.0
|
||||
import json
|
||||
@ -116,26 +116,30 @@ def cal_c1c2c3c4(param, heigt):
|
||||
|
||||
def gen_im_from_params(params, type="lines"):
|
||||
# 依据bim数据生成bim图
|
||||
# type: line points
|
||||
## 计算bim图长和宽
|
||||
max_y = -999999
|
||||
max_y_idx = -1
|
||||
max_x = -999999
|
||||
max_x_idx = -1
|
||||
# type: # line points
|
||||
## 确定整个bim图的长度和宽度边界
|
||||
max_y = -999999 # y坐标最大值
|
||||
max_y_idx = -1 # y坐标最大值的索引
|
||||
max_x = -999999 # x坐标最大值
|
||||
max_x_idx = -1 # x坐标最大值的索引
|
||||
|
||||
# 遍历,找到矩形x坐标最大值和矩形y坐标最大值
|
||||
for i, param in enumerate(params):
|
||||
# x中心点坐标加上宽度的一半 = 当前矩形的x最大坐标
|
||||
if param["x"] + param["w"] / 2 > max_x:
|
||||
# 如果是最大的x坐标就更新
|
||||
max_x = param["x"] + param["w"] / 2
|
||||
max_x_idx = i
|
||||
# y中心点坐标加上高度的一半 = 当前矩形的y最大坐标
|
||||
if param["center"] + param["h"] / 2 > max_y:
|
||||
# 如果是最大的y坐标就更新
|
||||
max_y = param["center"] + param["h"] / 2
|
||||
max_y_idx = i
|
||||
## 为了不贴边，x，y固定加 500
|
||||
FILL_VALUE = 0
|
||||
bim_width = int(max_x)
|
||||
bim_height = int(max_y)
|
||||
bim_width += FILL_VALUE
|
||||
bim_height += FILL_VALUE
|
||||
bim_channels = 1
|
||||
|
||||
padding_value = 0 # 内边距,避免整个bim图片贴着边缘展示
|
||||
bim_width = int(max_x) + padding_value
|
||||
bim_height = int(max_y) + padding_value
|
||||
bim_channels = 3
|
||||
im = np.zeros((bim_height, bim_width, bim_channels), dtype=np.uint8)
|
||||
|
||||
for i, param in enumerate(params):
|
||||
@ -145,7 +149,7 @@ def gen_im_from_params(params, type="lines"):
|
||||
[param['x2'], param['y2']],
|
||||
[param['x3'], param['y3']],
|
||||
[param['x4'], param['y4']]])
|
||||
cv2.polylines(im, [pts], True, (255,0,0), 8)
|
||||
cv2.polylines(im, [pts], True, (255,255,0), 8)
|
||||
elif type == "points":
|
||||
cv2.circle(im, (param['x1'], param['y1']), 1, 255, 20)
|
||||
cv2.circle(im, (param['x2'], param['y2']), 1, 255, 20)
|
||||
@ -155,9 +159,31 @@ def gen_im_from_params(params, type="lines"):
|
||||
|
||||
return im
|
||||
|
||||
def gen_points_from_params(params):
|
||||
|
||||
"""
|
||||
判断点是否在区域内部
|
||||
True 在
|
||||
False 不在
|
||||
如果没有提供区域,不做判断,返回True
|
||||
"""
|
||||
def is_inside_roi(point, roi_w, roi_h):
|
||||
if (roi_w != None and roi_h != None):
|
||||
x = point[0]
|
||||
y = point[1]
|
||||
if (x <= 0 or y <= 0 or y >= roi_h or x >= roi_w ):
|
||||
# 不在区域内部
|
||||
return False
|
||||
else:
|
||||
# 在区域内部
|
||||
return True
|
||||
else:
|
||||
# 没有提供区域,不做判断,默认返回True
|
||||
return True
|
||||
|
||||
|
||||
def gen_points_from_params(params,roi_w=None,roi_h=None):
|
||||
# 依据bim数据生成bim图
|
||||
# type: line points
|
||||
# type line points
|
||||
## 计算bim图长和宽
|
||||
max_y = -999999
|
||||
max_y_idx = -1
|
||||
@ -170,19 +196,26 @@ def gen_points_from_params(params):
|
||||
if param["center"] + param["h"] / 2 > max_y:
|
||||
max_y = param["center"] + param["h"] / 2
|
||||
max_y_idx = i
|
||||
## 为了不贴边，x，y固定加 500
|
||||
bim_height = int(max_y)
|
||||
|
||||
points = []
|
||||
adjacency = [] # 邻接矩阵
|
||||
for i, param in enumerate(params):
|
||||
cal_c1c2c3c4(param, bim_height)
|
||||
points.append([param['x1'], param['y1'], i])
|
||||
points.append([param['x2'], param['y2'], i])
|
||||
points.append([param['x3'], param['y3'], i])
|
||||
points.append([param['x4'], param['y4'], i])
|
||||
# points.append([param['x'], int((param['y3'] + param['y2']) / 2 )])
|
||||
points.append([param['x_center'], param['y_center'], i])
|
||||
# 过滤框框之外的点
|
||||
# 排序点 按照上左、上右、下右、下左的顺时针顺序
|
||||
if (roi_w == None and roi_h == None):
|
||||
cal_c1c2c3c4(param, bim_height)
|
||||
# 过滤点,把在roi区域之外的点全部过滤掉.
|
||||
if(is_inside_roi([param['x1'], param['y1']], roi_w, roi_h)):
|
||||
points.append([param['x1'], param['y1'], i])
|
||||
if(is_inside_roi([param['x2'], param['y2']], roi_w, roi_h)):
|
||||
points.append([param['x2'], param['y2'], i])
|
||||
if(is_inside_roi([param['x3'], param['y3']], roi_w, roi_h)):
|
||||
points.append([param['x3'], param['y3'], i])
|
||||
if(is_inside_roi([param['x4'], param['y4']], roi_w, roi_h)):
|
||||
points.append([param['x4'], param['y4'], i])
|
||||
if(is_inside_roi([param['x_center'], param['y_center']], roi_w, roi_h)):
|
||||
points.append([param['x_center'], param['y_center'], i])
|
||||
if (roi_w != None and roi_h != None):
|
||||
print(f"[经区域过滤之后的点数一共为] ====== [{len(points)}]")
|
||||
return points
|
||||
|
||||
def topological_similarity(adj1, adj2):
|
||||
@ -302,19 +335,12 @@ if __name__ == "__main__":
|
||||
# _im_edge_sobel = _sobel(sub_zero)
|
||||
# _, _im_thresh = cv2.threshold(_im_edge_sobel, 5, 255, cv2.THRESH_BINARY)
|
||||
# cnts = _findContours(_im_thresh)
|
||||
sub_im = cv2.imread("data_sub/test_1/wide_image.png")
|
||||
sub_im2 = cv2.imread("data_sub/test_1/wide_image.png")
|
||||
original_rectangle = read_from_json("data_sub/test_2/data_sub.json")
|
||||
# 过滤矩形
|
||||
# cnts 过滤之后的矩形
|
||||
# sub_im 裁剪之后的图像
|
||||
cnts,sub_im = filter_rectangle("data_sub/test_2/wide_image.png", original_rectangle)
|
||||
sub_zero = np.zeros_like(sub_im)
|
||||
cnts = read_from_json("data_sub/test_1/data_sub.json")
|
||||
# cnts的矩形画在sub_im 上
|
||||
for i in range(len(cnts)):
|
||||
p = cnts[i]
|
||||
cv2.rectangle(sub_im2, (p['x'], p['y']), (p['x'] + p['width'], p['y'] + p['height']), (0, 0, 255), 2)
|
||||
# 写编号
|
||||
cv2.putText(sub_im2, str(i), (p['x'], p['y']), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 0, 255), 2)
|
||||
cv2.imshow("sub_im_before_filter", sub_im2)
|
||||
# 过滤点
|
||||
cnts = filter_points("data_sub/test_1/wide_image.png",cnts)
|
||||
for contour in cnts:
|
||||
# x, y, w, h = cv2.boundingRect(contour)
|
||||
x = contour["x"]
|
||||
@ -323,12 +349,12 @@ if __name__ == "__main__":
|
||||
h = contour["height"]
|
||||
|
||||
# 由于定位框大小大于预埋件大小，因此这里需要做缩放处理
|
||||
k = int(h*0.01) # roi2
|
||||
k = int(h*0.01) # roi1
|
||||
x += k
|
||||
y += k
|
||||
w -= int(k) #
|
||||
h -= int(k) # 2k
|
||||
kh = int(h * 0.01) # roi1
|
||||
kw = int(w * 0.01) # roi1
|
||||
x += int(kw)
|
||||
y += int(kh)
|
||||
w -= int(kw)
|
||||
h -= int(kh)
|
||||
param = {}
|
||||
param["x1"] = x
|
||||
param["y1"] = y
|
||||
@ -403,7 +429,21 @@ if __name__ == "__main__":
|
||||
## 1.3计算所有点到该预埋件左上点的，点个数，平均极半径 和 平均极角度
|
||||
sum_r, sum_theta = 0.0,0
|
||||
count = 0
|
||||
pts = gen_points_from_params(_sub_sort_params)
|
||||
# 测试,画出所有的pts
|
||||
# for i, p in enumerate(_sub_sort_params):
|
||||
# # 画点
|
||||
# cv2.circle(sub_im, (p["x_center"], p["y_center"]), 2, (0, 255, 255), -1)
|
||||
# cv2.imshow("_sub_sort_params", sub_im)
|
||||
# cv2.waitKey(0)
|
||||
pts = gen_points_from_params(_sub_sort_params,sub_roi_width,sub_roi_height)
|
||||
# # 测试,画出所有的pts
|
||||
for i, p in enumerate(pts):
|
||||
# 画点
|
||||
cv2.circle(sub_im, (p[0], p[1]), 2, (0, 255, 255), -1)
|
||||
# 写编号
|
||||
cv2.putText(sub_im, str(i), (p[0], p[1]), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 0, 255), 1)
|
||||
cv2.imshow("sub_im_points", sub_im)
|
||||
# cv2.waitKey(0)
|
||||
polar_list = []
|
||||
for i, pt in enumerate(pts):
|
||||
r, theta = cartesian_to_polar(polar_origin_x, polar_origin_y, pt[0], pt[1])
|
||||
@ -414,6 +454,9 @@ if __name__ == "__main__":
|
||||
sum_r /= count * sub_roi_w_base_len
|
||||
sum_theta /= count
|
||||
print(count, sum_r, sum_theta)
|
||||
print(f"[所有点到该预埋件左上点的个数] ====== [{count}]")
|
||||
print(f"[所有点到该预埋件左上点的平均极半径] ====== [{sum_r}]")
|
||||
print(f"[所有点到该预埋件左上点的平均极角] ====== [{sum_theta}]")
|
||||
|
||||
# 初始化候选预埋件
|
||||
candi_params = []
|
||||
@ -428,7 +471,6 @@ if __name__ == "__main__":
|
||||
bim_im = gen_im_from_params(data_bim["params"])
|
||||
# sub_im = gen_im_from_params(data_sub["params"])# 需要读取
|
||||
min_match_score = 999999
|
||||
times = 0
|
||||
for i, param in enumerate(candi_params):
|
||||
tmp_roi_w_base_len = param['x2'] - param['x1']
|
||||
scale = tmp_roi_w_base_len / sub_roi_w_base_len
|
||||
@ -509,7 +551,7 @@ if __name__ == "__main__":
|
||||
print(f"起始预埋件ID：{id}，置信度为：{score[0]}，点数量：{score[1]}，平均长度为：{score[2]}，平均角度为：{score[3]}")
|
||||
print(f"match_sscore为：{match_score}")
|
||||
bim_im = cv2.rectangle(bim_im, param["start_point"], \
|
||||
param["end_point"], 255, 6)
|
||||
param["end_point"], 100 * (i+1), 6)
|
||||
|
||||
elapsed_time = time.time() - start_time
|
||||
print(f"Execution time: {elapsed_time:.4f} seconds")
|
||||
|
||||
84
utils.py
84
utils.py
@ -4,58 +4,92 @@ import logging
|
||||
import cv2
|
||||
|
||||
|
||||
"""
|
||||
根据裁剪之后的图片,每个点的坐标需要重新计算,以新的图片的宽高作为坐标系
|
||||
"""
|
||||
def re_cal_point(point,offset):
|
||||
# 相当于所有的x坐标向左平移了offset个距离
|
||||
point['x'] = point['x'] - offset
|
||||
|
||||
def filter_points(image_path,points):
|
||||
# 高度过大点过滤参数
|
||||
|
||||
|
||||
"""
|
||||
过滤矩形
|
||||
1 高度过大的不要
|
||||
2 整个矩形全部身体都在裁剪区域之外的不要
|
||||
|
||||
返回值:
|
||||
1 过滤之后的矩形
|
||||
2 裁剪之后的图片
|
||||
"""
|
||||
def filter_rectangle(image_path, points):
|
||||
# 高度过大矩形过滤参数
|
||||
max_height_rate = 0.5 # 矩形高度占整个画面高度的最大比例，如果超过该比例，则认为是无效矩形
|
||||
|
||||
# x范围过滤，只保留中间部分的点，只保留矩形完整出现在在画面中间的那部分矩形。
|
||||
left_x_cut_rate=0.2 # 左边界的裁剪比例，从左边开始裁剪百分之多少
|
||||
right_x_cut_rate=0.2 # 右边界的裁剪比例，从右边开始裁剪百分之多少
|
||||
# 裁剪参数
|
||||
left_x_cut_rate=0.15 # 左边界的裁剪比例，从左边开始裁剪百分之多少
|
||||
right_x_cut_rate=0.15 # 右边界的裁剪比例，从右边开始裁剪百分之多少
|
||||
image = cv2.imread(image_path)
|
||||
image_height = image.shape[0]
|
||||
image_width = image.shape[1]
|
||||
image_x_min = image_width * left_x_cut_rate
|
||||
image_x_max = image_width * (1 - right_x_cut_rate)
|
||||
image_height = image.shape[0] # 获取图片高度
|
||||
image_width = image.shape[1] # 获取图片宽度
|
||||
image_x_min = int(image_width * left_x_cut_rate) # 左边界的裁剪点
|
||||
image_x_max = int(image_width * (1 - right_x_cut_rate)) # 右边界的裁剪点
|
||||
|
||||
#开始过滤点
|
||||
#开始过滤矩形
|
||||
bad_point_index = []
|
||||
print(f'开始过滤点,原有点数为{len(points)}')
|
||||
print(f'开始过滤矩形,原有矩形数为{len(points)}')
|
||||
for index in range(len(points)):
|
||||
point = points[index]
|
||||
|
||||
# 高度过大过滤
|
||||
if point['height'] > image_height * max_height_rate:
|
||||
bad_point_index.append(index)
|
||||
continue
|
||||
|
||||
|
||||
# x坐标范围过滤
|
||||
# 原point中的x和y,是矩形左上角的坐标
|
||||
x_min = point['x'] # 矩形四个点坐标中x的最小值
|
||||
x_max = point['x'] + point['width'] # 矩形四个点坐标中x的最大值
|
||||
|
||||
# 如果矩形x的最小值小于左边界，去除这个矩形
|
||||
if x_min < image_x_min:
|
||||
# x坐标范围过滤,整个矩形全部身体都在裁剪区域之外的不要
|
||||
x_min = point['x'] # 矩形四个矩形坐标中x的最小值
|
||||
x_max = point['x'] + point['width'] # 矩形四个矩形坐标中x的最大值
|
||||
# 如果矩形x的 最大值 小于 左边界，去除这个矩形
|
||||
if x_max < image_x_min:
|
||||
bad_point_index.append(index)
|
||||
continue
|
||||
# 如果矩形x的最大值大于右边界，去除这个矩形
|
||||
if x_max > image_x_max:
|
||||
# 如果矩形x的 最小值 大于 右边界，去除这个矩形
|
||||
if x_min > image_x_max:
|
||||
bad_point_index.append(index)
|
||||
continue
|
||||
|
||||
# 删除bad_point_index
|
||||
# 过滤,只保留有效矩形
|
||||
filtered_points = []
|
||||
for i, point in enumerate(points):
|
||||
# 如果当前矩形的索引在bad_point_index中，则去除这个矩形
|
||||
if i not in bad_point_index:
|
||||
# 重新计算点的坐标
|
||||
re_cal_point(point,image_x_min)
|
||||
# 塞入结果
|
||||
filtered_points.append(point)
|
||||
print(f'过滤点结束,过滤之后的点数为{len(filtered_points)}')
|
||||
return filtered_points
|
||||
print(f'过滤矩形结束,过滤之后的矩形数为{len(filtered_points)}')
|
||||
|
||||
# 图片裁剪
|
||||
# 裁剪图片 (height方向不变，宽度方向裁剪)
|
||||
cropped_image = image[:, image_x_min:image_x_max]
|
||||
# 展示
|
||||
# cv2.imshow("cropped_image", cropped_image)
|
||||
# cv2.imshow("image", image)
|
||||
# for i in range(len(filtered_points)):
|
||||
# p = filtered_points[i]
|
||||
# cv2.rectangle(cropped_image, (p['x'], p['y']), (p['x'] + p['width'], p['y'] + p['height']), (0, 0, 255), 2)
|
||||
# # 写编号
|
||||
# cv2.putText(cropped_image, str(i), (p['x'], p['y']), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 0, 255), 2)
|
||||
# cv2.imshow("cropped_image_draw", cropped_image)
|
||||
# cv2.waitKey(0)
|
||||
return filtered_points, cropped_image
|
||||
|
||||
|
||||
|
||||
# 测试代码
|
||||
# def read_from_json(file_path):
|
||||
# with open(file_path, 'r') as f:
|
||||
# loaded_array = json.load(f)
|
||||
# return loaded_array
|
||||
# cnts = read_from_json("data_sub/test_1/data_sub.json")
|
||||
# filter_points("data_sub/test_1/wide_image.png",cnts)
|
||||
# filter_rectangle("data_sub/test_1/wide_image.png",cnts)
|
||||
Loading…
Reference in New Issue
Block a user