Line sketch images of real objects in a scene may contain junctions that represent corners of objects. The work reported in the paper shows how measures of the angles at a junction provide information concerning the structure of physical corners. The authors concentrate on the corners of a cube and derive the probability view densities of the corner angles under isotropic positioning of the viewpoint. Interpretation of the results shows that non-general viewing positions are quite likely, and that such placements of the viewpoint should not be treated as pathological when interpreting line sketches or analyzing scenes.<>
{"title":"Junction view densities in images","authors":"R. Malik, T. Whangbo","doi":"10.1109/IAI.1994.336668","DOIUrl":"https://doi.org/10.1109/IAI.1994.336668","url":null,"abstract":"Line sketch images of real objects in a scene may contain junctions that represent corners of objects. The work reported in the paper shows how measures of the angles at a junction provide information concerning the structure of physical corners. The authors concentrate on the corners of a cube and derive the probability view densities of the corner angles under isotropic positioning of the viewpoint. Interpretation of the results shows that non-general viewing positions are quite likely, and that such placements of the viewpoint should not be treated as pathological when interpreting line sketches or analyzing scenes.<<ETX>>","PeriodicalId":438137,"journal":{"name":"Proceedings of the IEEE Southwest Symposium on Image Analysis and Interpretation","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128502927","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A. Nabout, R. Gerhards, B. Su, H. A. Nour Eldin, W. Kuhbauch
The automatic identification of plant species is a great challenge because their patterns are complex and uncertain. In this paper, the fuzzy set theory was applied to identify weed species. A membership function was established. The experiment has shown, that the average rate of correct identification has improved from 67% to greater than 82%.<>
{"title":"Plant species identification using fuzzy set theory","authors":"A. Nabout, R. Gerhards, B. Su, H. A. Nour Eldin, W. Kuhbauch","doi":"10.1109/IAI.1994.336686","DOIUrl":"https://doi.org/10.1109/IAI.1994.336686","url":null,"abstract":"The automatic identification of plant species is a great challenge because their patterns are complex and uncertain. In this paper, the fuzzy set theory was applied to identify weed species. A membership function was established. The experiment has shown, that the average rate of correct identification has improved from 67% to greater than 82%.<<ETX>>","PeriodicalId":438137,"journal":{"name":"Proceedings of the IEEE Southwest Symposium on Image Analysis and Interpretation","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125585419","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
D. S. Kang, Norman C. Griswold, Nasser Kehtarnavaz
One of the most noteworthy problems associated with conventional pattern recognition methods is that it is not easy to extract feature vectors from images which are not translation, rotation, and scale change invariant in outdoor noisy environments. This paper describes the development of an invariant traffic sign recognition system capable of tolerating the above variations. The signs are restricted to three types of warning signs and are all of red color. The developed method is insensitive to brightness changes as well as invariant to translation, rotation, scale change, and noise. The architecture of this system is based upon neural network supervised learning after geometrical transformations have been applied. The performance of this system is compared with other invariant approaches in terms of the percentage of correct decisions in outdoor noisy environments.<>
{"title":"An invariant traffic sign recognition system based on sequential color processing and geometrical transformation","authors":"D. S. Kang, Norman C. Griswold, Nasser Kehtarnavaz","doi":"10.1109/IAI.1994.336679","DOIUrl":"https://doi.org/10.1109/IAI.1994.336679","url":null,"abstract":"One of the most noteworthy problems associated with conventional pattern recognition methods is that it is not easy to extract feature vectors from images which are not translation, rotation, and scale change invariant in outdoor noisy environments. This paper describes the development of an invariant traffic sign recognition system capable of tolerating the above variations. The signs are restricted to three types of warning signs and are all of red color. The developed method is insensitive to brightness changes as well as invariant to translation, rotation, scale change, and noise. The architecture of this system is based upon neural network supervised learning after geometrical transformations have been applied. The performance of this system is compared with other invariant approaches in terms of the percentage of correct decisions in outdoor noisy environments.<<ETX>>","PeriodicalId":438137,"journal":{"name":"Proceedings of the IEEE Southwest Symposium on Image Analysis and Interpretation","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123145243","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A multiresolution analysis (MRA) of L/sup 2/(R/sup 3/), which supports an orthogonal wavelet decomposition and a quadrature mirror filter pyramid algorithm, is the framework for estimating image motion. Instead of the usual iterative pyramid algorithm, a fast, direct derivation of coarse resolution space-time images within the MRA is developed. Two spatio-temporal filtering methods are described. The orientation of surfaces in R/sup 3/ gives robust indication of image motion. However, the alternative method with oriented texture energies in the spatio-temporal image fares poorly when velocities vary.<>
{"title":"Motion estimation using the multiresolution analysis of L/sup 2/(R/sup 3/)","authors":"R. L. Allen","doi":"10.1109/IAI.1994.336671","DOIUrl":"https://doi.org/10.1109/IAI.1994.336671","url":null,"abstract":"A multiresolution analysis (MRA) of L/sup 2/(R/sup 3/), which supports an orthogonal wavelet decomposition and a quadrature mirror filter pyramid algorithm, is the framework for estimating image motion. Instead of the usual iterative pyramid algorithm, a fast, direct derivation of coarse resolution space-time images within the MRA is developed. Two spatio-temporal filtering methods are described. The orientation of surfaces in R/sup 3/ gives robust indication of image motion. However, the alternative method with oriented texture energies in the spatio-temporal image fares poorly when velocities vary.<<ETX>>","PeriodicalId":438137,"journal":{"name":"Proceedings of the IEEE Southwest Symposium on Image Analysis and Interpretation","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131833183","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
When a vision sensor is used to track an object in an outdoor realistic navigational environment, it is subjected to unexpected movements or vibrations of the mounting platform. In this paper, the performance of such a setup in terms of range and heading angle errors is studied. The noise generated by the navigational environment is represented in two ways: sensor rotation angle errors and image coordinates errors. A consistent detectable region is obtained such that the tracked object is always seen by the sensor. Based on this region, a reliable region consisting of no singularity points is defined so that the range error does not become infinity. The optimum values of a controllable subspace, consisting of the object height and depression angle, with respect to an uncontrollable subspace, consisting of object coordinates and sensor movement errors, are then found by employing the mini-max estimator for the worst case performance and the minimum mean-squared estimator for the average performance.<>
{"title":"Analysis of sensor movement errors in monocular vision-based tracking systems","authors":"W. Sohn, N. Kehtarnavaz","doi":"10.1109/IAI.1994.336678","DOIUrl":"https://doi.org/10.1109/IAI.1994.336678","url":null,"abstract":"When a vision sensor is used to track an object in an outdoor realistic navigational environment, it is subjected to unexpected movements or vibrations of the mounting platform. In this paper, the performance of such a setup in terms of range and heading angle errors is studied. The noise generated by the navigational environment is represented in two ways: sensor rotation angle errors and image coordinates errors. A consistent detectable region is obtained such that the tracked object is always seen by the sensor. Based on this region, a reliable region consisting of no singularity points is defined so that the range error does not become infinity. The optimum values of a controllable subspace, consisting of the object height and depression angle, with respect to an uncontrollable subspace, consisting of object coordinates and sensor movement errors, are then found by employing the mini-max estimator for the worst case performance and the minimum mean-squared estimator for the average performance.<<ETX>>","PeriodicalId":438137,"journal":{"name":"Proceedings of the IEEE Southwest Symposium on Image Analysis and Interpretation","volume":"109 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128593841","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper presents a fast and reliable vergence control method using a hierarchical image structure for active vision systems. In this method, the sign pyramid is generated from the Laplacian pyramid. The search for the common fixation point and the disparity value of that point is performed by sign-correlation matching and a coarse-to-fine search strategy. This method works well for a very large range of disparities and is not sensitive to calibration problems common to stereo images.<>
{"title":"Vergence control using a hierarchical image structure","authors":"Changhoon Yim, A. Bovik","doi":"10.1109/IAI.1994.336672","DOIUrl":"https://doi.org/10.1109/IAI.1994.336672","url":null,"abstract":"This paper presents a fast and reliable vergence control method using a hierarchical image structure for active vision systems. In this method, the sign pyramid is generated from the Laplacian pyramid. The search for the common fixation point and the disparity value of that point is performed by sign-correlation matching and a coarse-to-fine search strategy. This method works well for a very large range of disparities and is not sensitive to calibration problems common to stereo images.<<ETX>>","PeriodicalId":438137,"journal":{"name":"Proceedings of the IEEE Southwest Symposium on Image Analysis and Interpretation","volume":"12 11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121283816","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The Lorenz (1905) information measure (LIM) is a function of the observed probability sequence of digital signals, similar to the signal entropy, and is approximately linearly related to the mean absolute error (MAE) in simulations employing uncorrupted and corrupted 2-dimensional Gaussian and magnetic resonance (MR) images. Unlike the MAE, the LIM does not require an uncorrupted reference signal for a distance computation. However, for the particular difference signal case imposed by the definition of the MAE, the LIM is asymptotically bounded by the MAE/signal quantization number ratio. Therefore, in applications where an uncorrupted signal does not exist, and thus, the MAE is undefined, the LIM provides a comparable signal processing performance measure.<>
{"title":"Theoretical and experimental comparison of the Lorenz information measure, entropy, and the mean absolute error","authors":"T. McMurray, J. Pearce","doi":"10.1109/IAI.1994.336688","DOIUrl":"https://doi.org/10.1109/IAI.1994.336688","url":null,"abstract":"The Lorenz (1905) information measure (LIM) is a function of the observed probability sequence of digital signals, similar to the signal entropy, and is approximately linearly related to the mean absolute error (MAE) in simulations employing uncorrupted and corrupted 2-dimensional Gaussian and magnetic resonance (MR) images. Unlike the MAE, the LIM does not require an uncorrupted reference signal for a distance computation. However, for the particular difference signal case imposed by the definition of the MAE, the LIM is asymptotically bounded by the MAE/signal quantization number ratio. Therefore, in applications where an uncorrupted signal does not exist, and thus, the MAE is undefined, the LIM provides a comparable signal processing performance measure.<<ETX>>","PeriodicalId":438137,"journal":{"name":"Proceedings of the IEEE Southwest Symposium on Image Analysis and Interpretation","volume":"17 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1994-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116710158","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：