-
Notifications
You must be signed in to change notification settings - Fork 1
/
morph_features.m
198 lines (160 loc) · 4.82 KB
/
morph_features.m
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
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
% Dec. 10 2018 - Changed Fourier descriptors to be middle descriptors, not
% first 10 - Patrick Leo
function [feats] = morph_features(x,y)
% Function to extract a set of morphological features.
% Input:
% [x,y] Input coordinates of the boundary points.
%
% Output:
% feats Feature vector.
%
% The features are arranged in the following manner:
% Order the coordinates
xy = order([x,y]');
% Get the centroid and area
[xc,yc,area]=centroid(xy);
x = xy(:,1);
y = xy(:,2);
% % Plot the polygons and the centroids
% figure;
% scatter(x,-y,'r');
% hold on;
% plot(xc,-yc,'o');
distance=sqrt((x-xc).^2+(y-yc).^2);
dist_min=min(distance);
dist_max=max(distance);
% Draw circle from centroid to max radius
theta = 0:.01:2*pi;
x_c1 = zeros(1,length(theta));
y_c1 = zeros(1,length(theta));
for i=1:length(theta)
x_c1(i)=dist_max*cos(theta(i))+xc;
y_c1(i)=-dist_max*sin(theta(i))-yc;
end
% hold on; plot(x_c1,y_c1,'k');
% Draw circle from centroid to min radius
theta = 0:.01:2*pi;
x_c2 = zeros(1,length(theta));
y_c2 = zeros(1,length(theta));
for i=1:length(theta)
x_c2(i)=dist_min*cos(theta(i))+xc;
y_c2(i)=dist_min*sin(theta(i))+yc;
end
% hold on; plot(x_c2,y_c2,'k')
%==========================================================================
% Get variance and standard deviation
var = cov(distance);
stdv = std(distance);
%==========================================================================
%get maximum area and Area Ratio
max_area = pi*dist_max^2;
Area_Ratio = area/max_area;
%==========================================================================
%Ratio between average distance and maximum distance
dist_mean = mean(distance);
Dist_Ratio = dist_mean/dist_max;
%==========================================================================
%normalizing distance to find the variance and std
dists = distance/dist_max;
dists_std = std(dists);
dists_var = cov(dists);
%==========================================================================
%new distance ratio defined
dratio = distratio(xy);
%==========================================================================
%area to perimeter ratio
[paratio,peri] = periarea(xy,area);
% Will be infinite if area is 0, as in case of straight line
if(paratio == Inf)
% warning('Infinite perimiter to area ratio');
paratio = nan;
end
%==========================================================================
%the SMOOTHNESS METRIC
D = distance;
s = length(D);
sm = zeros(1,length(D));
for i=1:s;
if i==1;
sm(i) = abs(D(i)-((D(i+1)+D(s))/2));
else if i==s;
sm(i) = abs(D(i)-((D(1)+D(i-1))/2));
else if i>=2 && i<=s-1
sm(i) = abs(D(i)-((D(i+1)+D(i-1))/2));
end
end
end
end
smooth = sum(sm);
%==========================================================================
% Fourier Descriptors of boundary
% z = frdescp([x,y]);
% fd = z(1:10);
z = frdescpUncentered([x,y]);
% First descriptor is DC component (which is translationally variant)
if(length(z) < 11)
z(end+1:11) = nan;
end
% divide by sum(z) for scale invariance
fd = z(2:11)/nansum(z);
%============
% Invariant moments
B = bound2im([x,y]);
phi = invmoments(B);
%===========================
% Get the fractal dimension
% Will be infinite if two identical coordinates have one coordinate between
% them
frac_dim = fractal_dim([x,y]);
if(frac_dim == Inf)
frac_dim = nan;
end
%==========================================
% Para are all the features that are extracted.
feats = [Area_Ratio;
Dist_Ratio;
dists_std;
dists_var;
dratio;
paratio;
smooth;
phi';
frac_dim;
fd];
%para=[Area_Ratio;Dist_Ratio;dists_std;dists_var];
% fid = fopen(name,'w');
% fprintf(fid,'%15.10f\n',para);
% fprintf(fid,'%15.10f \t %15.10f\n',[x,y]);
% fclose(fid);
%to get all the features for glands and get the mean automatically.
% files=['g1.xls';'g2.xls'; 'g3.xls'; 'g4.xls'];
% for i=1:size(files,1);
% fid = fopen(files(i,:));
% feature(i,:) = fscanf(fid,'%g',[1 115]); %
% fclose(fid)
% end
% feature=feature';
% mean_feature=mean(feature,2);
% mean_feature=mean_feature';
% save mean_feature mean_feature
% % old one
% %CAVE =[0.4999;0.7178;0.1419;0.0211];
% CAVE = [0.47542;0.70094; 0.14158; 0.021017; 0.76584];
%
% % old one
% %BAVE =[0.2308;0.5183;0.1984;0.0450];
% BAVE = [0.27297;0.58401; 0.20965; 0.045246; 0.63293];
%
% n1=norm(para-CAVE);
% n2=norm(para-BAVE);
%
% if n1 < n2
% disp('============================================')
% disp('===============Sorry, Cancerous=============')
% disp('~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~')
% else if n2 < n1
% disp('========================================')
% disp('=================Benign=================')
% disp('~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~')
% end
% end