-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathvis_tracks_single_wnearGT.py
141 lines (111 loc) · 5.55 KB
/
vis_tracks_single_wnearGT.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
'''
visulize randomly several single tracks
'''
import glob
import pandas as pd
import matplotlib.pyplot as plt
import os
import skimage.io as io
import seaborn as sns
import random
import argparse
from utils import readXML, find_near
import numpy as np
__author__ = "Yudong Zhang"
palette = sns.color_palette('hls', 30)
def get_color(seed):
random.seed(seed)
# random color
bbox_color = random.choice(palette)
bbox_color = [int(255 * c) for c in bbox_color][::-1]
cl='#'+hex(bbox_color[0])[-2:]+hex(bbox_color[1])[-2:]+hex(bbox_color[2])[-2:]
cl = cl.upper()
return cl
def xml2df(xmlfilepath):
poslist = readXML(xmlfilepath) # x, y, t, z, float(p)
P = [np.array(_) for _ in poslist]
M = np.vstack(P)
detection_total = pd.DataFrame(M[:,[0,1,2,4]])
detection_total.columns=['pos_x','pos_y','frame','trackid']
return detection_total
def parse_args_():
parser = argparse.ArgumentParser()
parser.add_argument('--imgfolder', type=str, default='/data/ldap_shared/synology_shared/zyd/data/20220611_detparticle/challenge/MICROTUBULE snr 7 density low')
parser.add_argument('--trackcsvpath', type=str, default='./prediction/20240301_15_25_56/track_result.csv')
parser.add_argument('--vis_save', type=str, default='./prediction/20240301_15_25_56/track_vis')
parser.add_argument('--img_fmt', type=str, default='**t{:03d}**.tif')
parser.add_argument('--vis_dot', default=False, action='store_true' )
parser.add_argument('--vistrack_number', default=50, type=int)
parser.add_argument('--vistrack_length', type=int, default=2)
parser.add_argument('--GTtrackxmlpath', type=str, default='./dataset/tracks10/GTxml/test_2024_04_08__14_44_25.xml')
parser.add_argument('--visGT_near_num',type=int, default=5)
parser.add_argument('--vispast_length',type=int, default=None)
opt = parser.parse_args()
return opt
if __name__ == '__main__':
opt = parse_args_()
result_pa = opt.trackcsvpath
imgfolder = opt.imgfolder
savefolder = opt.vis_save
os.makedirs(savefolder, exist_ok=True)
result = pd.read_csv(result_pa,header=0)
filename = result_pa.split('/')[-1].replace('.csv','')
print('[Info] Start')
GT_df = xml2df(opt.GTtrackxmlpath)
alltracks_idx = list(set(result['trackid']))
alltracks_idx.sort()
viscount = 0
for idx_, the_id in enumerate(alltracks_idx):
print(f"[Info] Processing {idx_}/{len(alltracks_idx)}")
this_idtrack = result[result['trackid'] == the_id]
if len(this_idtrack) >= opt.vistrack_length:
viscount += 1
print(f"[Info] Finish {viscount}/{opt.vistrack_number}")
print(f"[Info] Processing {idx_}/{len(alltracks_idx)}")
if viscount >= opt.vistrack_number:
break
this_idtrack = this_idtrack.sort_values('frame')
for fr in range(int(this_idtrack['frame'].values.min()), int(this_idtrack['frame'].values.max()+1)):
imgpath = glob.glob(os.path.join(imgfolder,opt.img_fmt.format(fr)))
assert len(imgpath) == 1
img = io.imread(imgpath[0])
H,W = img.shape
plt.figure()
plt.imshow(img,'gray')
plt.axis('off')
# ID_color = get_color(the_id)
ID_color = 'r'
# this_iddet = result[result['trackid']==the_id].sort_values('frame')
this_iddet_near = this_idtrack[(this_idtrack['frame']<=fr)] #&(this_iddet['frame']>fr-10)
this_iddet_near = this_iddet_near.sort_values(by='frame')
xlist = [max(min(x, W-2),1) for x in this_iddet_near['pos_x']]
ylist = [max(min(y, H-2),1) for y in this_iddet_near['pos_y']]
plt.plot(xlist,ylist,linewidth=0.5,color=ID_color)
if opt.vis_dot:
plt.scatter([xlist[-1]],[ylist[-1]],color=ID_color, marker='o', edgecolors=ID_color, s=1,linewidths=1)
# vis near GT
thisframe_gtdf = GT_df[GT_df['frame'] == fr]
nearp = find_near(thisframe_gtdf, x = xlist[-1],y = ylist[-1])
numnear = min(opt.visGT_near_num, nearp.shape[0])
for i in range(numnear):
onenearid = nearp[i,-2]
ID_color = get_color(onenearid) if onenearid != the_id else get_color(onenearid+1)
its_tracklet = GT_df[GT_df['trackid'] == onenearid]
past_tracklet = its_tracklet[its_tracklet['frame'] <= fr]
past_tracklet = past_tracklet.sort_values(by='frame')
xlist = [max(min(x, W-2),1) for x in past_tracklet['pos_x']]
ylist = [max(min(y, H-2),1) for y in past_tracklet['pos_y']]
if opt.vispast_length:
if len(xlist)>opt.vispast_length:
xlist = xlist[-opt.vispast_length:]
ylist = ylist[-opt.vispast_length:]
plt.plot(xlist,ylist,'--',linewidth=0.5,color=ID_color)
if opt.vis_dot:
plt.scatter([xlist[-1]],[ylist[-1]],color=ID_color, marker='x', edgecolors=ID_color, s=0.7,linewidths=0.7)
plt.savefig(os.path.join(savefolder, 'track%d_%03d.jpg'%(the_id,fr)), bbox_inches='tight',dpi=300,pad_inches=0.0)
plt.close()
else:
# print(f'track{the_id} has only one length')
pass
# break
print('[Info] Success!')