The goal of FLIR image enhancement is to obtain a good display by simultaneously compressing the global scene dynamic range and enhancing the local area contrast. Six candidates for FLIR image enhancement are presente...
详细信息
The goal of FLIR image enhancement is to obtain a good display by simultaneously compressing the global scene dynamic range and enhancing the local area contrast. Six candidates for FLIR image enhancement are presented and some experimental results shown. The six enhancement candidates are: (1) variable threshold zonal filtering, (2) statistical differencing operator, (3) unsharp masking, (4) prototype automatic target screener technique, (5) constant variance, and (6) histogram equalization. An experiment using subjective criteria to judge the enhanced image quality was conducted. The results show that the variable threshold zonal filter, prototype automatic target screener, and unsharp masking methods were the superior techniques.
Simple statisticalmethods for enhancing PCM and DPCM monochrome pictures corrupted by transmission bit errors are described. For PCM signals the enhancement is achieved by operating exclusively on the corrupted image...
详细信息
Simple statisticalmethods for enhancing PCM and DPCM monochrome pictures corrupted by transmission bit errors are described. For PCM signals the enhancement is achieved by operating exclusively on the corrupted image, without recourse to any side-information from the transmitter. The DPCM signals have simple protection code words that increase the bit-rate by 12 percent (if majority protection coding is used). Photographs of the original, corrupted and enhanced images are presented, together with SNR as a function of percentage bit-error-rate (BER) for the corrupted and enhanced picture. For 0. 1 less than BER less than 1. 0, the SNR gains for PCM and DPCM are approximately equals 10 and 6 dB, respectively. For BER less than 0. 05 percent there is imperceptible picture degradation.
Forty-six papers are presented in this special conference issue. They are arranged broadly in correspondence with the following sequence of sessions: medical applications, line drawing processing, statisticalmethods,...
详细信息
Forty-six papers are presented in this special conference issue. They are arranged broadly in correspondence with the following sequence of sessions: medical applications, line drawing processing, statisticalmethods, scene analysis and data structure, theoretical problems and clustering, applications, shape analysis and object location, syntactic methods, microscopic imageprocessing, acoustic signal recognition and perceptual modelling, edge detection, image segmentation and texture, special hardware and systems, picture transformation and recognition and character recognition. All papers are abstracted separately. Technical and professional papers from this conference are indexed with the conference code no. 00432 in the Ei Engineering Meetings (TM) database produced by Engineering Information, Inc.
Given two "nonlinear filtering problems" described by the processes dx (t)i = fi (xi(t)) dt+gi (xi(t))dwi(t) i=1,2, dx (t)i = hi (xi(t)) dt+dvi(t), we define a notion of strong equivalence relating the solut...
详细信息
Given two "nonlinear filtering problems" described by the processes dx (t)i = fi (xi(t)) dt+gi (xi(t))dwi(t) i=1,2, dx (t)i = hi (xi(t)) dt+dvi(t), we define a notion of strong equivalence relating the solutions to the corresponding Mortensen-Zakai equations dui (t,x) = Lui i(t,x)dt + Li iui (t,x)dyt i, i=1,2, which allows solution of one problem to be obtained easily from solutions of the other. We give a geometric picture of this equivalence as a group of local transformations acting on manifolds of solutions. We then show that by knowing the full invariance group of the time invariant equations dui (t,x) = Lui i (t,x)dt, i=1,2, we can analyze strong equivalence for the filtering problems. In particular if the two time invariant parabolic operators are in the same orbit of the invariance group we can show strong equivalence for the filtering problems. As a result filtering problems are separated into equivalent classes which correspond to orbits of invariance groups of parabolic operators. As specific example we treat V. BeneŠ's case establishing from this point of view the necessity of the Riccati equation.
The three-dimensional shape analysis problem is a very demanding test of shape analysis algorithms. Previous approaches to the problem have employed global features such as moments and Fourier descriptors. Global feat...
详细信息
The three-dimensional shape analysis problem is a very demanding test of shape analysis algorithms. Previous approaches to the problem have employed global features such as moments and Fourier descriptors. Global features lack the capacity for solving the partial shape recognition problem, in which only part of the unknown shape is available. Previous approaches to local shape analysis have employed structural (syntactic) methods, but these methods have so far failed to solve the three-dimensional problem. This study describes a hybrid structural/statistical local shape analysis algorithm which is applied to the three-dimensional problem.
The problem of estimating the error probability of a given classification system is considered. statistical properties of the empirical error count (C) and the average conditional error (R) estimators are studied. It ...
详细信息
The problem of estimating the error probability of a given classification system is considered. statistical properties of the empirical error count (C) and the average conditional error (R) estimators are studied. It is shown that in the large sample case the R estimator is unbiased and its variance is less than that of the C estimator. In contrast to conventional methods of Bayes error estimation the unbiasedness of the R estimator for a given classifier can be obtained only at the price of an additional set of classified samples. On small test sets the R estimator may be subject to a pessimistic bias caused by the averaging phenomenon characterising the functioning of conditional error estimators.
The three-dimensional shape analysis problem is a very demanding test of shape analysis algorithms. Previous approaches to the problem have employed global features such as moments and Fourier descriptors. Global feat...
详细信息
The three-dimensional shape analysis problem is a very demanding test of shape analysis algorithms. Previous approaches to the problem have employed global features such as moments and Fourier descriptors. Global features lack the capacity for solving the partial shape recognition problem, in which only part of the unknown shape is available. Previous approaches to local shape analysis have employed structural (syntactic) methods, but these methods have so far failed to solve the three-dimensional problem. This study describes a hybrid structural/statistical local shape analysis algorithm which is applied to the three-dimensional problem.
The problem of estimating the error probability of a given classification system is considered. statistical properties of the empirical error count (C) and the average conditional error (R) estimators are studied. It ...
详细信息
The problem of estimating the error probability of a given classification system is considered. statistical properties of the empirical error count (C) and the average conditional error (R) estimators are studied. It is shown that in the large sample case the R estimator is unbiased and its variance is less than that of the C estimator. In contrast to conventional methods of Bayes error estimation the unbiasedness of the R estimator for a given classifier can be obtained only at the price of an additional set of classified samples. On small test sets the R estimator may be subject to a pessimistic bias caused by the averaging phenomenon characterising the functioning of conditional error estimators.
The proceedings contain 153 papers. The topics discussed include: software development: a failure and a success;an analysis of the performance of a software development methodology;effective computer program developme...
The proceedings contain 153 papers. The topics discussed include: software development: a failure and a success;an analysis of the performance of a software development methodology;effective computer program development and use;a methodology for the development of reliable synchronization software;a relational DBMS conforming to an architecture which incorporates a physical navigation language;task driven image understanding: lisp programming for vision research;some new algorithms and software implementation methods for pattern research recognition;statisticalimageprocessing and recognition;reflection on the implementation of a software design;functional multiprocessing in an experimental digital switching office;and a microprocessor line concentration system.
暂无评论