1 /*M///////////////////////////////////////////////////////////////////////////////////////
3 // IMPORTANT: READ BEFORE DOWNLOADING, COPYING, INSTALLING OR USING.
5 // By downloading, copying, installing or using the software you agree to this license.
6 // If you do not agree to this license, do not download, install,
7 // copy or use the software.
10 // Intel License Agreement
12 // Copyright (C) 2000, Intel Corporation, all rights reserved.
13 // Third party copyrights are property of their respective owners.
15 // Redistribution and use in source and binary forms, with or without modification,
16 // are permitted provided that the following conditions are met:
18 // * Redistribution's of source code must retain the above copyright notice,
19 // this list of conditions and the following disclaimer.
21 // * Redistribution's in binary form must reproduce the above copyright notice,
22 // this list of conditions and the following disclaimer in the documentation
23 // and/or other materials provided with the distribution.
25 // * The name of Intel Corporation may not be used to endorse or promote products
26 // derived from this software without specific prior written permission.
28 // This software is provided by the copyright holders and contributors "as is" and
29 // any express or implied warranties, including, but not limited to, the implied
30 // warranties of merchantability and fitness for a particular purpose are disclaimed.
31 // In no event shall the Intel Corporation or contributors be liable for any direct,
32 // indirect, incidental, special, exemplary, or consequential damages
33 // (including, but not limited to, procurement of substitute goods or services;
34 // loss of use, data, or profits; or business interruption) however caused
35 // and on any theory of liability, whether in contract, strict liability,
36 // or tort (including negligence or otherwise) arising in any way out of
37 // the use of this software, even if advised of the possibility of such damage.
41 #include "precomp.hpp"
42 #include "opencv2/video/tracking_c.h"
44 /*======================= KALMAN FILTER =========================*/
45 /* State vector is (x,y,w,h,dx,dy,dw,dh). */
46 /* Measurement is (x,y,w,h). */
48 /* Dynamic matrix A: */
49 const float A8[] = { 1, 0, 0, 0, 1, 0, 0, 0,
50 0, 1, 0, 0, 0, 1, 0, 0,
51 0, 0, 1, 0, 0, 0, 1, 0,
52 0, 0, 0, 1, 0, 0, 0, 1,
53 0, 0, 0, 0, 1, 0, 0, 0,
54 0, 0, 0, 0, 0, 1, 0, 0,
55 0, 0, 0, 0, 0, 0, 1, 0,
56 0, 0, 0, 0, 0, 0, 0, 1};
58 /* Measurement matrix H: */
59 const float H8[] = { 1, 0, 0, 0, 0, 0, 0, 0,
60 0, 1, 0, 0, 0, 0, 0, 0,
61 0, 0, 1, 0, 0, 0, 0, 0,
62 0, 0, 0, 1, 0, 0, 0, 0};
64 /* Matrices for zero size velocity: */
65 /* Dinamic matrix A: */
66 const float A6[] = { 1, 0, 0, 0, 1, 0,
73 /* Measurement matrix H: */
74 const float H6[] = { 1, 0, 0, 0, 0, 0,
83 class CvBlobTrackPostProcKalman:public CvBlobTrackPostProcOne
92 float m_DataNoiseSize;
95 CvBlobTrackPostProcKalman();
96 ~CvBlobTrackPostProcKalman();
97 CvBlob* Process(CvBlob* pBlob);
99 virtual void ParamUpdate();
100 }; /* class CvBlobTrackPostProcKalman */
103 CvBlobTrackPostProcKalman::CvBlobTrackPostProcKalman()
105 m_ModelNoise = 1e-6f;
106 m_DataNoisePos = 1e-6f;
107 m_DataNoiseSize = 1e-1f;
110 m_DataNoiseSize *= (float)pow(20.,2.);
112 m_DataNoiseSize /= (float)pow(20.,2.);
115 AddParam("ModelNoise",&m_ModelNoise);
116 AddParam("DataNoisePos",&m_DataNoisePos);
117 AddParam("DataNoiseSize",&m_DataNoiseSize);
120 m_pKalman = cvCreateKalman(STATE_NUM,4);
121 memcpy( m_pKalman->transition_matrix->data.fl, A, sizeof(A));
122 memcpy( m_pKalman->measurement_matrix->data.fl, H, sizeof(H));
124 cvSetIdentity( m_pKalman->process_noise_cov, cvRealScalar(m_ModelNoise) );
125 cvSetIdentity( m_pKalman->measurement_noise_cov, cvRealScalar(m_DataNoisePos) );
126 CV_MAT_ELEM(*m_pKalman->measurement_noise_cov, float, 2,2) = m_DataNoiseSize;
127 CV_MAT_ELEM(*m_pKalman->measurement_noise_cov, float, 3,3) = m_DataNoiseSize;
128 cvSetIdentity( m_pKalman->error_cov_post, cvRealScalar(1));
129 cvZero(m_pKalman->state_post);
130 cvZero(m_pKalman->state_pre);
132 SetModuleName("Kalman");
135 CvBlobTrackPostProcKalman::~CvBlobTrackPostProcKalman()
137 cvReleaseKalman(&m_pKalman);
140 void CvBlobTrackPostProcKalman::ParamUpdate()
142 cvSetIdentity( m_pKalman->process_noise_cov, cvRealScalar(m_ModelNoise) );
143 cvSetIdentity( m_pKalman->measurement_noise_cov, cvRealScalar(m_DataNoisePos) );
144 CV_MAT_ELEM(*m_pKalman->measurement_noise_cov, float, 2,2) = m_DataNoiseSize;
145 CV_MAT_ELEM(*m_pKalman->measurement_noise_cov, float, 3,3) = m_DataNoiseSize;
148 CvBlob* CvBlobTrackPostProcKalman::Process(CvBlob* pBlob)
150 CvBlob* pBlobRes = &m_Blob;
152 CvMat Zmat = cvMat(4,1,CV_32F,Z);
157 m_pKalman->state_post->data.fl[0+4] = CV_BLOB_X(pBlob)-m_pKalman->state_post->data.fl[0];
158 m_pKalman->state_post->data.fl[1+4] = CV_BLOB_Y(pBlob)-m_pKalman->state_post->data.fl[1];
161 m_pKalman->state_post->data.fl[2+4] = CV_BLOB_WX(pBlob)-m_pKalman->state_post->data.fl[2];
162 m_pKalman->state_post->data.fl[3+4] = CV_BLOB_WY(pBlob)-m_pKalman->state_post->data.fl[3];
164 m_pKalman->state_post->data.fl[0] = CV_BLOB_X(pBlob);
165 m_pKalman->state_post->data.fl[1] = CV_BLOB_Y(pBlob);
166 m_pKalman->state_post->data.fl[2] = CV_BLOB_WX(pBlob);
167 m_pKalman->state_post->data.fl[3] = CV_BLOB_WY(pBlob);
170 { /* Nonfirst call: */
171 cvKalmanPredict(m_pKalman,0);
172 Z[0] = CV_BLOB_X(pBlob);
173 Z[1] = CV_BLOB_Y(pBlob);
174 Z[2] = CV_BLOB_WX(pBlob);
175 Z[3] = CV_BLOB_WY(pBlob);
176 cvKalmanCorrect(m_pKalman,&Zmat);
177 cvMatMulAdd(m_pKalman->measurement_matrix, m_pKalman->state_post, NULL, &Zmat);
178 CV_BLOB_X(pBlobRes) = Z[0];
179 CV_BLOB_Y(pBlobRes) = Z[1];
180 // CV_BLOB_WX(pBlobRes) = Z[2];
181 // CV_BLOB_WY(pBlobRes) = Z[3];
187 void CvBlobTrackPostProcKalman::Release()
192 static CvBlobTrackPostProcOne* cvCreateModuleBlobTrackPostProcKalmanOne()
194 return (CvBlobTrackPostProcOne*) new CvBlobTrackPostProcKalman;
197 CvBlobTrackPostProc* cvCreateModuleBlobTrackPostProcKalman()
199 return cvCreateBlobTrackPostProcList(cvCreateModuleBlobTrackPostProcKalmanOne);
201 /*======================= KALMAN FILTER =========================*/
205 /*======================= KALMAN PREDICTOR =========================*/
206 class CvBlobTrackPredictKalman:public CvBlobTrackPredictor
210 CvBlob m_BlobPredict;
214 float m_DataNoisePos;
215 float m_DataNoiseSize;
218 CvBlobTrackPredictKalman();
219 ~CvBlobTrackPredictKalman();
221 void Update(CvBlob* pBlob);
222 virtual void ParamUpdate();
227 }; /* class CvBlobTrackPredictKalman */
230 void CvBlobTrackPredictKalman::ParamUpdate()
232 cvSetIdentity( m_pKalman->process_noise_cov, cvRealScalar(m_ModelNoise) );
233 cvSetIdentity( m_pKalman->measurement_noise_cov, cvRealScalar(m_DataNoisePos) );
234 CV_MAT_ELEM(*m_pKalman->measurement_noise_cov, float, 2,2) = m_DataNoiseSize;
235 CV_MAT_ELEM(*m_pKalman->measurement_noise_cov, float, 3,3) = m_DataNoiseSize;
238 CvBlobTrackPredictKalman::CvBlobTrackPredictKalman()
240 m_ModelNoise = 1e-6f;
241 m_DataNoisePos = 1e-6f;
242 m_DataNoiseSize = 1e-1f;
245 m_DataNoiseSize *= (float)pow(20.,2.);
247 m_DataNoiseSize /= (float)pow(20.,2.);
250 AddParam("ModelNoise",&m_ModelNoise);
251 AddParam("DataNoisePos",&m_DataNoisePos);
252 AddParam("DataNoiseSize",&m_DataNoiseSize);
255 m_pKalman = cvCreateKalman(STATE_NUM,4);
256 memcpy( m_pKalman->transition_matrix->data.fl, A, sizeof(A));
257 memcpy( m_pKalman->measurement_matrix->data.fl, H, sizeof(H));
259 cvSetIdentity( m_pKalman->process_noise_cov, cvRealScalar(m_ModelNoise) );
260 cvSetIdentity( m_pKalman->measurement_noise_cov, cvRealScalar(m_DataNoisePos) );
261 CV_MAT_ELEM(*m_pKalman->measurement_noise_cov, float, 2,2) = m_DataNoiseSize;
262 CV_MAT_ELEM(*m_pKalman->measurement_noise_cov, float, 3,3) = m_DataNoiseSize;
263 cvSetIdentity( m_pKalman->error_cov_post, cvRealScalar(1));
264 cvZero(m_pKalman->state_post);
265 cvZero(m_pKalman->state_pre);
267 SetModuleName("Kalman");
270 CvBlobTrackPredictKalman::~CvBlobTrackPredictKalman()
272 cvReleaseKalman(&m_pKalman);
275 CvBlob* CvBlobTrackPredictKalman::Predict()
279 cvKalmanPredict(m_pKalman,0);
280 m_BlobPredict.x = m_pKalman->state_pre->data.fl[0];
281 m_BlobPredict.y = m_pKalman->state_pre->data.fl[1];
282 m_BlobPredict.w = m_pKalman->state_pre->data.fl[2];
283 m_BlobPredict.h = m_pKalman->state_pre->data.fl[3];
285 return &m_BlobPredict;
288 void CvBlobTrackPredictKalman::Update(CvBlob* pBlob)
291 CvMat Zmat = cvMat(4,1,CV_32F,Z);
292 m_BlobPredict = pBlob[0];
296 m_pKalman->state_post->data.fl[0+4] = CV_BLOB_X(pBlob)-m_pKalman->state_post->data.fl[0];
297 m_pKalman->state_post->data.fl[1+4] = CV_BLOB_Y(pBlob)-m_pKalman->state_post->data.fl[1];
300 m_pKalman->state_post->data.fl[2+4] = CV_BLOB_WX(pBlob)-m_pKalman->state_post->data.fl[2];
301 m_pKalman->state_post->data.fl[3+4] = CV_BLOB_WY(pBlob)-m_pKalman->state_post->data.fl[3];
303 m_pKalman->state_post->data.fl[0] = CV_BLOB_X(pBlob);
304 m_pKalman->state_post->data.fl[1] = CV_BLOB_Y(pBlob);
305 m_pKalman->state_post->data.fl[2] = CV_BLOB_WX(pBlob);
306 m_pKalman->state_post->data.fl[3] = CV_BLOB_WY(pBlob);
309 { /* Nonfirst call: */
310 Z[0] = CV_BLOB_X(pBlob);
311 Z[1] = CV_BLOB_Y(pBlob);
312 Z[2] = CV_BLOB_WX(pBlob);
313 Z[3] = CV_BLOB_WY(pBlob);
314 cvKalmanCorrect(m_pKalman,&Zmat);
317 cvKalmanPredict(m_pKalman,0);
323 CvBlobTrackPredictor* cvCreateModuleBlobTrackPredictKalman()
325 return (CvBlobTrackPredictor*) new CvBlobTrackPredictKalman;
327 /*======================= KALMAN PREDICTOR =========================*/