queso-0.57.1
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Pages
brute.cpp
Go to the documentation of this file.
1 //----------------------------------------------------------------------
2 // File: brute.cpp
3 // Programmer: Sunil Arya and David Mount
4 // Description: Brute-force nearest neighbors
5 // Last modified: 05/03/05 (Version 1.1)
6 //----------------------------------------------------------------------
7 // Copyright (c) 1997-2005 University of Maryland and Sunil Arya and
8 // David Mount. All Rights Reserved.
9 //
10 // This software and related documentation is part of the Approximate
11 // Nearest Neighbor Library (ANN). This software is provided under
12 // the provisions of the Lesser GNU Public License (LGPL). See the
13 // file ../ReadMe.txt for further information.
14 //
15 // The University of Maryland (U.M.) and the authors make no
16 // representations about the suitability or fitness of this software for
17 // any purpose. It is provided "as is" without express or implied
18 // warranty.
19 //----------------------------------------------------------------------
20 // History:
21 // Revision 0.1 03/04/98
22 // Initial release
23 // Revision 1.1 05/03/05
24 // Added fixed-radius kNN search
25 //----------------------------------------------------------------------
26 
27 #include <ANN/ANNx.h> // all ANN includes
28 #include "pr_queue_k.h" // k element priority queue
29 
30 //----------------------------------------------------------------------
31 // Brute-force search simply stores a pointer to the list of
32 // data points and searches linearly for the nearest neighbor.
33 // The k nearest neighbors are stored in a k-element priority
34 // queue (which is implemented in a pretty dumb way as well).
35 //
36 // If ANN_ALLOW_SELF_MATCH is ANNfalse then data points at distance
37 // zero are not considered.
38 //
39 // Note that the error bound eps is passed in, but it is ignored.
40 // These routines compute exact nearest neighbors (which is needed
41 // for validation purposes in ann_test.cpp).
42 //----------------------------------------------------------------------
43 
44 ANNbruteForce::ANNbruteForce( // constructor from point array
45  ANNpointArray pa, // point array
46  int n, // number of points
47  int dd) // dimension
48 {
49  dim = dd; n_pts = n; pts = pa;
50 }
51 
52 ANNbruteForce::~ANNbruteForce() { } // destructor (empty)
53 
54 void ANNbruteForce::annkSearch( // approx k near neighbor search
55  ANNpoint q, // query point
56  int k, // number of near neighbors to return
57  ANNidxArray nn_idx, // nearest neighbor indices (returned)
58  ANNdistArray dd, // dist to near neighbors (returned)
59  double eps) // error bound (ignored)
60 {
61  ANNmin_k mk(k); // construct a k-limited priority queue
62  int i;
63 
64  if (k > n_pts) { // too many near neighbors?
65  annError("Requesting more near neighbors than data points", ANNabort);
66  }
67  // run every point through queue
68  for (i = 0; i < n_pts; i++) {
69  // compute distance to point
70  ANNdist sqDist = annDist(dim, pts[i], q);
71  if (ANN_ALLOW_SELF_MATCH || sqDist != 0)
72  mk.insert(sqDist, i);
73  }
74  for (i = 0; i < k; i++) { // extract the k closest points
75  dd[i] = mk.ith_smallest_key(i);
76  nn_idx[i] = mk.ith_smallest_info(i);
77  }
78 }
79 
80 int ANNbruteForce::annkFRSearch( // approx fixed-radius kNN search
81  ANNpoint q, // query point
82  ANNdist sqRad, // squared radius
83  int k, // number of near neighbors to return
84  ANNidxArray nn_idx, // nearest neighbor array (returned)
85  ANNdistArray dd, // dist to near neighbors (returned)
86  double eps) // error bound
87 {
88  ANNmin_k mk(k); // construct a k-limited priority queue
89  int i;
90  int pts_in_range = 0; // number of points in query range
91  // run every point through queue
92  for (i = 0; i < n_pts; i++) {
93  // compute distance to point
94  ANNdist sqDist = annDist(dim, pts[i], q);
95  if (sqDist <= sqRad && // within radius bound
96  (ANN_ALLOW_SELF_MATCH || sqDist != 0)) { // ...and no self match
97  mk.insert(sqDist, i);
98  pts_in_range++;
99  }
100  }
101  for (i = 0; i < k; i++) { // extract the k closest points
102  if (dd != NULL)
103  dd[i] = mk.ith_smallest_key(i);
104  if (nn_idx != NULL)
105  nn_idx[i] = mk.ith_smallest_info(i);
106  }
107 
108  return pts_in_range;
109 }
ANNpoint * ANNpointArray
Definition: ANN.h:376
ANNpointArray pts
Definition: ANN.h:541
double ANNdist
Definition: ANN.h:159
ANNcoord * ANNpoint
Definition: ANN.h:375
int n_pts
Definition: ANN.h:540
int dim
Definition: ANN.h:539
int annkFRSearch(ANNpoint q, ANNdist sqRad, int k=0, ANNidxArray nn_idx=NULL, ANNdistArray dd=NULL, double eps=0.0)
Definition: brute.cpp:80
Definition: ANNx.h:48
~ANNbruteForce()
Definition: brute.cpp:52
DLL_API ANNdist annDist(int dim, ANNpoint p, ANNpoint q)
Definition: ANN.cpp:46
ANNdist * ANNdistArray
Definition: ANN.h:377
ANNbruteForce(ANNpointArray pa, int n, int dd)
Definition: brute.cpp:44
void annkSearch(ANNpoint q, int k, ANNidxArray nn_idx, ANNdistArray dd, double eps=0.0)
Definition: brute.cpp:54
ANNidx * ANNidxArray
Definition: ANN.h:378
void annError(const char *msg, ANNerr level)
Definition: ANN.cpp:169
const ANNbool ANN_ALLOW_SELF_MATCH
Definition: ANN.h:235

Generated on Tue Jun 5 2018 19:48:54 for queso-0.57.1 by  doxygen 1.8.5