I read papers and I understand what kinds of features that GIFT extracts but the problem I am having is about the unpacking files that are created when I extract features. Here is a good example: I have a .fts file and when I read it I get something like this 63:108 63:123 63:132 63:178 63:188 63:197 220:14 15:1 15:1 15:1 15:1 16:1 16:1 16:1 17:1 17:1 17:1 1:0 1:0 1:0 128:63 128:63 128:63 128:63 128:63 128:63 63:89 63:99 63:109 63:220 230:22 0:0 113:23 123:23 224:23 0:0 63:89 63:97 63:107 63:35 63:46 63:56 63:67 63:77 63:87 63:97 63:106 63:117 63:127 63:137 63:147 63:157 63:168 63:177 63:187 63:55 63:65 63:75 63:85 63:145 63:156 63:49 63:59 63:159 63:167 63:1 63:145 63:205 215:31 0:0 0:0 0:0 0:0 0:0 0:0 0:0 0:0 128:63 1:0 1:0 1:0 1:0 1:0 1:0 1:0 0:0 128:63 34:1 34:1 34:1 34:1 34:1 34:1 34:1 34:1 34:1 34:1 1:0 0:0 0:0 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 64:38 143:38 81:40 242:40 5:41 135:41 165:41 175:41 71:42 79:42 92:42 109:42 119:42 129:42 230:42 0:128 43:1 43:1 43:1 43:1 43:1 0:0 0:128 57:44 77:44 87:44 97:44 107:44 117:44 177:44 37:45 81:45 189:45 199:45 46:1 46:1 46:1 46:1 46:1 0:0 6:47 89:47 120:47 129:47 156:47 167:47 179:47 249:47 0:0 0:0 0:0 0:0 0:0 0:0 0:0 0:128 63:49 63:57 63:87 63:123 63:133 63:137 63:147 63:197 212:49 1:0 0:0 26:50 37:50 106:50 118:50 125:50 141:50 155:50 187:50 197:50 1:0 0:0 63:8 63:19 63:26 63:45 63:57 63:92 63:102 63:107 63:115 63:157 63:168 63:176 63:185 63:207 216:51 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 63:229 0:0 3:53 13:53 25:53 60:53 71:53 84:53 120:53 124:53 133:53 143:53 176:53 188:53 220:53 0:0 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 63:207 222:54 55:1 55:1 55:1 55:1 0:128 0:128 0:128 0:128 0:128 0:128 0:128 0:128 128:63 63:63 63:87 63:97 63:127 63:198 207:59 0:128 0:128 0:128 0:128 0:128 0:128 0:128 0:128 61:1 61:1 61:1 61:1 61:1 61:1 1:0 0:0 0:0 0:0 0:0 0:0 0:0 0:0 0:128 128:63 239:62 4:63 43:63 117:63 125:63 133:63 183:63 78:64 90:64 127:64 237:64 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 63:63 63:93 63:47 63:251 0:0 0:0 0:0 0:0 0:0 0:128 128:63 49:72 59:72 79:72 90:72 100:72 114:72 127:72 134:72 210:72 1:0 0:0 63:253 0:0 0:0 0:0 0:0 0:0 0:0 0:0 0:0 0:0 128:63 243:73 0:0 0:0 0:0 0:0 0:0 0:0 0:0 0:0 0:0 0:0 0:0 0:128 128:63 235:74 0:128 75:1 75:1 75:1 75:1 76:1 76:1 77:1 77:1 77:1 1:0 0:0 0:0 0:0 0:0 0:0 0:0 63:254 0:128 0:128 0:128 0:128 0:128 128:63 128:63 128:63 128:63 128:63 128:63 128:63 128:63 245:82 63:6 63:14 63:24 63:33 63:147 63:157 63:163 63:175 63:47 63:59 63:80 63:87 63:127 63:179 63:188 63:217 21:85 30:85 85:1 1:0 0:129 0:193 201:200 128:59 60:40 1:0 1:0 0:161 0:129 0:193 193:192 176:48 136:136 128:0 61:85 70:85 71:85 0:137 0:225 192:60 1:0 128:0 93:63 176:61 16:61 60:90 63:91 61:92 61:93 94:85 85:1 85:1 1:0 0:129 0:217 209:208 48:61 128:59 128:59 12:63 28:62 61:113 85:1 85:1 85:1 0:177 0:225 160:32 68:63 20:62 0:61 124:85 125:85 126:85 130:85 131:85 1:0 1:0 1:0 1:0 1:0 0:225 128:128 200:61 141:85 142:85 143:85 144:85 145:85 147:85
I can see that the first three digits are for Feature Id and the second part is frequency (Thanks for explanation). Now how do I go from here to know what feature ID is (144 or 145). I tried to read the code about Features extraction but I am not sure why I am missing how I can connect this kind of output to Inverted Files. Please let me know if I am missing something. Sometimes is not easy to fully understand another person's code. My automate goal to be able to know how many features I extracted from the image and what they are. I hope I am not being a pain but I am sure this info is somewhere in this code but I am missing it. Even just pseudocode on how this is done might help. Just out of curiosity, does it matter that my collection is not colored? Thanks again Epa Uwimana Quoting Henning Müller <[EMAIL PROTECTED]>: > Hi, > > the papers expalin pretty well what kind of features exist. Then you > can > create a list of feature number vs exact feature description as > Wolfgang > explained in his mail . The executable to do so is write_feature_desc > or > something like that in the FeatureExtraction directory and it > requires > as input a ppm image of the size 256x256. Have you tried this? > > Cheers, Henning > > Epa wrote: > > I read some papers about GIFT back in 2000 and 1999 and I still > can't find > > what values stand for HSV, GABOR, COLORMAP or SIZE from the out of > this script > > > > > > > > > > > > > > _______________________________________________ > > help-GIFT mailing list > > [email protected] > > http://lists.gnu.org/mailman/listinfo/help-gift > > > > -- > --------------------------------------- > Henning Müller (PhD) > University & Hospitals of Geneva > Service of Medical Informatics > 24, rue Micheli-du-Crest > CH-1211 Geneva 14, Switzerland > Tel +41 22 372-6175 > Fax +41 22 372-8680 > email [EMAIL PROTECTED] > > _______________________________________________ help-GIFT mailing list [email protected] http://lists.gnu.org/mailman/listinfo/help-gift
