23.6 Unsupervised Filter
23.6.1 Matching
“Unsupervised /Matching/1:1 (Left vs Right)” Menu Item
Use the “Unsupervised /Matching/1:1 (Left vs Right)” menu item to make a 1:1 Matching. 1:1 Matching compares the key with the image displayed in the right picture box. The “Key” button selects the key image. The right image can be selected by “F”, “>”, “<” buttons.
“Unsupervised/Matching/1:N (key vs a1.txt)” Menu Item
Use the “Unsupervised/Matching/1:N (key vs a1.txt)” menu item to make a 1:N Matching. 1:N Matching compares the key with the images in a1.txt. The “Key” button selects the key image. 1:N Matching requires the images in the search source being converted into signatures in advance and saved in a1.txt.
To make an 1:N Matching:
o Click the “Key” button, and select an image;
o Click “Unsupervised/Matching/1:N (key vs a1.txt)”.
The results are in file, b1.txt, which will be opened at the end of the computation.
“Unsupervised/Matching/N:N Match (a1.txt vs a1.txt)” Menu Item
Use the “Unsupervised/Matching/N:N Match (a1.txt vs a1.txt)” menu item to make an N:N Matching. N: N Matching compares each image, in a1.txt, with every image in a1.txt. N:N Matching requires the images in the search source being converted into signatures in advance and saved in a1.txt. The results are in file, b1.txt, which will be opened at the end of the computation.
23.6.2 Analysis
“Unsupervised/N:M Analysis/Check (b1_matchlist.txt required)” Menu Item
Use the “Unsupervised/N:M Analysis/Check (b1_matchlist.txt required)” menu item to check the matching results. If this is a test run (i.e., you know the correct answers), you can see the matching results in seconds. To test the results in b1.txt, you must prepare b1_matchlist.txt file, which indicates the matching pairs. Once b1.txt and b1_matchlist.txt are prepared, this command will let you know your matching results in seconds.
“Unsupervised/N:M Analysis/Report (b1_matchlist.txt required)” Menu Item
Use the “Unsupervised/N:M Analysis/Check (b1_matchlist.txt required)” menu item to get a matched list and an unmatched list. The matched list contains those pairs in b1_matchlist.txt file that have been found by the ImageFinder. The unmatched list contains those pairs in b1_matchlist.txt file that have been missed by the ImageFinder.
“Unsupervised/N:M Analysis/Threhold (b1_matchlist.txt required)” Menu Item
Use the “Unsupervised/N:M Analysis/Threhold (b1_matchlist.txt required)” menu item to get the recommended threshold, based on b1_matchlist.txt file. This menu item will first find the matched list. From the matched list, it will find the minimum score and recommend it to you as the threshold. The output looks like this:
Possible Threshold = x
Absolute Threshold = y
The Possible Threshold is the threshold.
The Absolute Threshold specifies the score if an image only matches itself and no other images.
23.6.3 Results
“Unsupervised/Results/b1.txt” Menu Item
Use the “Unsupervised/Results/b1.txt” menu item to open b1.txt, which is the file containing the last matching result in text format.
“Unsupervised/Results/b1.htm” Menu Item
Use the “Unsupervised/Results/b1.htm” menu item to open b1.htm, which is the file containing the last matching result in html format.
“Unsupervised/Results/b1.html” Menu Item
Use the “Unsupervised/Results/b1.html” menu item to open b1.html, which is the file containing the last matching result in html format, with both input images and output images displayed in the web page.
23.6.4 Parameters
Unsupervised Parameter/Fault Tolerance Scale
Use the Fault Tolerance Scale textbox to control the amount of output. This parameter ranges from 0 to 100. The larger this number is, the more matches you will get. To set this parameter, enter a number between 0 and 100 to the text box.
Unsupervised/Parameter/Blurring Text Box
Use the Blurring Text Box to control the amount of output. "0%" Blurring means the exact match. When the "Blurring" is increased, you will get more and more similar images. As the Blurring goes higher, the speed will be slower. The Blurring settings range from 0 – 50. To set the Blurring, enter a number between 0 and 50 to the text box.
Unsupervised/Parameter/Sensitivity Text Box
Use the Sensitivity Text Box to adjust the search segment size. The Sensitivity parameter ranges from 0 (least sensitive) to 100 (most sensitive).
- To search small segment(s), use a high sensitivity search.
- To search large segment(s), use low sensitivity search.
The higher the parameter is set, the more results you will get. To set the Sensitivity, enter a number between 0 and 100 the text box. The default setting is 50.
Unsupervised/Parameter/Threshold Text Box
Use the Threshold to eliminate those retrieved images with the weights below a certain value. The result of image comparison is a "score", indicating the degree to which a match exists. This score is then compared to a pre-set Threshold to determine whether or not to declare a match.
To decide what threshold to use, you should make a test run first and look at the scores.
- Matching images have higher scores;
- Unmatched images have lower scores.
Select a threshold to separate these two groups. There will be a few images in the middle, representing both groups. Under these circumstances, the threshold selection depends on your application. In general, it is better to give no answer than a wrong answer. Assume you are searching images and all similar images have weights ranging from 1,000 to 10,000. It is possible that some other images will pop up with weights ranging from 10 to 100. To eliminate these images, you can set the External Cut to 1,000.
Unsupervised/Parameter/Relative Score Text Box
Use the “Relative Score” textbox to switch between two settings: the first setting gives the score in the range (0 – 100) and the second setting gives a score in the range (0 to maximum value).
Unsupervised/Parameter/Show File Text Box
Use the “Show File” textbox to decide whether you want the output file, b1.txt, to be displayed at the end of a matching.
Copyright (c) 1998 - 2006 Attrasoft, Inc. All rights reserved.