Changes To FLCC
Not logged in
     1         -<meta http-equiv="Refresh" content="5; url="> 
            1  +<title>FLCC Home Page</title>
            2  +
            3  +<p align="center">
            4  +<font size="5">
            5  +<b>FLCC:</b>
            6  +<i>Fast computation of Local Correlation Coefficients</i>
            7  +</font>
            8  +</p>
            9  +
           10  +
           11  +<h3>Why Use FLCC?</h3>
           12  +
           13  +<table border="0" cellpadding="1" bgcolor="#558195" align="right" hspace="10">
           14  +<tr><td>
           15  +<table border="0" cellpadding="10" bgcolor="white">
           16  +<tr><td>
           17  +<h3>Quick Links</h3>
           18  +<ul>
           19  +<li> [../doc/FLCC_Manual_v1.0.pdf | Doc]
           20  +<li> [../tar/flcc_v1.0.tgz | Download]
           21  +<li> [../COPYRIGHT-BSD2.txt | License]
           22  +<li> [./ | FAQ]
           23  +</ul>
           24  +</td></tr>
           25  +<tr><td>
           26  +<center><img src="fossil3.gif"></center>
           27  +</td></tr>
           28  +</table>
           29  +</table>
           30  +
           31  +<p>
           32  +The FLCC Library (Fast LCC) is a programming tool for fast computation
           33  +of Local Correlation Coefficients (LCCs) between an image (or a stream
           34  +of images) and an image template. The LCC computation is a basic
           35  +image-based information processing step which finds numerous
           36  +applications in areas such as template or pattern matching, image
           37  +registration, motion detection and many more. However, the LCC
           38  +computation has always been considered of high arithmetic complexity and
           39  +time consuming, especially for real-time applications, thus making the
           40  +usage of LCCs rather troublesome.
           41  +
           42  +<p>
           43  +This library intends to overcome this problem and provide users with a
           44  +simple yet powerful interface for computing the LCCs. Most
           45  +implementations so far tried to reduce computation time by sacrificing
           46  +the local normalization characteristic of the LCCs or approximating the
           47  +result in other lossy ways. This library, though, manages to reduce
           48  +computation time to a minimum, yet without making any compromises on the
           49  +quality of the result. The user can be sure that the output result is
           50  +accurately the “real” LCC distribution in any case, as exactly it is
           51  +defined formally.
           52  +
           53  +<p>
           54  +The performance benefit is achieved by two routes. On the one hand, the
           55  +library implements a set of optimized fast algorithms, selecting the
           56  +most appropriate for each case. On the other hand, it fully exploits
           57  +current top-notch architectures, namely multicore CPU processors and
           58  +manycore GPU devices. In other words, the library takes advantage of the
           59  +system computational resources, executing in parallel on multiple CPU
           60  +threads or having the work load carried out by the powerful GPU devices
           61  +of the system (according to what it deems to be faster).
           62  +