scispace - formally typeset
Search or ask a question

Showing papers on "Change detection published in 1989"


Journal ArticleDOI
TL;DR: An evaluation of results indicates that various procedures of change detection produce different maps of change even in the same environment.
Abstract: A variety of procedures for change detection based on comparison of multitemporal digital remote sensing data have been developed. An evaluation of results indicates that various procedures of change detection produce different maps of change even in the same environment.

3,361 citations


Journal ArticleDOI
TL;DR: Two techniques for change detection are presented that have been developed to deal with the more general scenario where illuination is not assumed to be constant, and results are presented for applying each of the techniques discussed to various image pairs.
Abstract: Change detection plays a very important role in many vision applications Most change detection algorithms assume that the illumination on a scene will remain constant Unfortunately, this assumption is not necessarily valid outside a well-controlled laboratory setting The accuracy of existing algorithms diminishes significantly when confronted with image sequences in which the illumination is allowed to vary In this note, we present two techniques for change detection that have been developed to deal with the more general scenario where illuination is not assumed to be constant A detailed description of both new methods, the derivative model method and the shading model method, is provided Results are presented for applying each of the techniques discussed to various image pairs

224 citations


Journal ArticleDOI
P. Strobach1
TL;DR: The results of this research indicate that a constrained joint process model involving only a single gain parameter and a shift parameter is the best tradeoff between performance and computational complexity.
Abstract: A summary is presented of a study on two-dimensional linear prediction models for image sequence processing and its application to change detection and scene coding. The study focused on two-dimensional joint process modeling of interframe relationships, the derivation of computationally efficient matching algorithms, and the implementation of a block-adaptive interframe predictor for use in interframe predictive coding and change detection. In the approach presented, the spatial nonstationarity is handled by an underlying quadtree segmentation structure. A maximum-likelihood criterion and a simpler minimum-variance criterion are discussed as detection and segmentation rules. The results of this research indicate that a constrained joint process model involving only a single gain parameter and a shift parameter is the best tradeoff between performance and computational complexity. >

32 citations


Proceedings ArticleDOI
23 May 1989
TL;DR: The authors address the problem of motion detection in an image sequence from the variations in time of the intensity distribution by the joint treatment of the detection of temporal changes and the reconstruction of mobile object masks according to a probabilistic formulation.
Abstract: The authors address the problem of motion detection in an image sequence from the variations in time of the intensity distribution. The problem is not limited to change detection but encompasses the recovery of the projections of moving areas in the image. The approach is characterized by the joint treatment of the detection of temporal changes and the reconstruction of mobile object masks according to a probabilistic formulation. More formally, spatio-temporal contextual information is introduced through Markovian models, using Gibbs distributions defined on a spatio-temporal neighborhood system. Then the problem at hand is stated as a statistical labeling one. To decide whether or not a point belongs to a moving area is equivalent to assigning to it a given label. A solution to this labeling problem is formulated according to the maximum a posteriori (MAP) criterion. Experiments with a real image sequence have been carried out. >

23 citations


Journal ArticleDOI
TL;DR: Results of this study indicate that the procedures which employ squared cusum and nondiscounted cusu of innovations yield smaller rates of false detection, and the two best change-detection statistics appear to possess good potential for industrial and business forecast monitoring applications.

15 citations


Patent
10 Jul 1989
TL;DR: In this article, an inter-picture distance between patterns was used to detect a broadcast program causing a scene change at a prescribed time interval by using an inter picture distance between pattern, detecting the scene change and detecting the time from scene change to the succeeding scene change.
Abstract: PURPOSE:To detect a broadcast program causing a scene change at a prescribed time interval surely by using an inter-picture distance between patterns, detecting the scene change and detecting the time from the scene change to the succeeding scene change CONSTITUTION:A television signal processing section 4 has a scene change detection section 5 and a commercial broadcast extraction processing section 6 The section 5 detects a scene change and the section 6 applies commercial broadcast extraction processing The scene change detection section 5 calculates the inter-picture distance between adjacent patterns of the television broadcast to detect the scene change section included in the television broadcast Moreover, the commercial broadcast extraction processing section 6 detects the time between the patterns included in the television broadcast and detects it to be a prescribed broadcast program when the time interval is coincident with a prescribed time

8 citations


Journal ArticleDOI
TL;DR: In this paper, the authors used profile analysis to discriminate changes in the shape of the power spectrum of complex acoustic signals, which is called profile analysis, and showed that the most critical feature of profile analysis is that detection of a change at one spectral location depends on energy at very remote spectral locations.
Abstract: Data are reported on the ability of listeners to discriminate changes in the power spectra of complex acoustic signals. Since randomizing both the phase spectrum and the overall level of the signal on each presentation does not change detection performance, it is known that detection is not based on the signal's phase spectrum or the absolute level of energy at a particular frequency location. The discrimination must be based on a change in the shape of the power spectrum of the signal, what is called profile analysis. The most critical feature of profile analysis is that detection of a change at one spectral location depends on energy at very remote spectral locations. For example, the detection of an increment in the intensity of a component at 1000 Hz is influenced by the presence or absence of energy more than 2 oct away. Recent experiments show that these effects may depend on whether the standard power spectra are fixed or changing over successive trials, since spectra with randomly changing levels do not show such effects. Finally, some recent experiments will be discussed that provide estimates of the spectral weights employed by profile listeners, using a technique developed by Dr. Bruce Berg. [This research was supported by NIH and AFOSR.]

4 citations


01 Jan 1989
TL;DR: This note presents two new methods for change detection: the derivative model method and the shading model method that have been developed to deal with the more general scenario where illuination is not assumed to be constant.
Abstract: Change detection plays a very important role in many vision applications. Most change detection algorithms assume that the illumination on a scene will remain constant. Unfortunately, this assumption is not necessarily valid outside a well-controlled laboratory setting. The accuracy of existing algorithms diminishes significantly when confronted with image sequences in which the illumination is allowed to vary. In this note, we present two techniques for change detection that have been developed to deal with the more general scenario where illuination is not assumed to be constant. A detailed description of both new methods, the derivative model method and the shading model method, is provided. Results are presented for applying each of the techniques discussed to various image pairs. Detecting changes is fundamental to one’s perception of the world. After all, the world we live in is dynamic, and the inputs to our senses are constantly changing. Many models have been proposed for detection of motion in the human visual system [13-171. In this note our concern is only change detection; motion detection may use change detection, but it is considered beyond the scope of this discussion. It is our aim is to develop robust change detection techniques for machine vision systems. It is not surprising that the process of change detection is fundamental to many machine vision applications. Systems that track moving objects [18], analyze cloud motion, mointor the growth of crops, or analyze traffic flow [2-41, are just a few examples of machine vision systems that use change detection algorithms. These algorithms provide the low level information that can be used by higher level algorithms to determine the information desired (the trajectory of an object, the growth of a tree, etc.). Therefore, for these systems to operate successfully, it is extremely important that change detection algorithms to accurate and robust. Change detection may take place either at the pixel level or at a higher level-by comparing features. We will address change detection at the pixel level. Our motivation in developing robust techniques for change detection at the pixel level is the possibility of very fast change detection for robotic applications. In this note, we present two new methods for change detection: the derivative model method and the shading model method. The derivative model method uses partial derivatives with respect to the pixel coordinates of a second order gray level surface model to compare regions and determine if a change has taken place. This

3 citations


Patent
21 Feb 1989
TL;DR: In this article, the authors proposed to improve the tracking performance with respect to the transient state by providing a transient change detection section detecting a signal transient change and an input/output deviation detection section detected the increased deviation in the signal and allowing a processing section to vary the time constant at the detection of transient change.
Abstract: PURPOSE:To improve the tracking performance with respect to the transient state by providing a transient change detection section detecting a signal transient change and an input/output deviation detection section detecting the increased deviation in the signal and allowing a processing section to vary the time constant at the detection of transient change. CONSTITUTION:A transient change detection section 2 obtains a difference between an input signal and an output signal between an input section 1 and a filter processing section 3 and judges it as the occurrence of a transient change if the difference exceeds a reference value. Moreover, an input/output deviation detection section 4 decided it as a large deviation if the integration value of the difference between the input signal and the output signal exceeds a reference value. A processing section 3 has a function having a smooth processing capability in the steady-state but changing the time constant sequentially at the entrance of a detection signal, and then can follow well a rapid transient change as well as a slow transient change. Thus, the titled digital filter acts stably in an alarm device or a control system where a transient change is to be caught surely.

2 citations


Patent
30 Mar 1989
TL;DR: In this article, the condition change detection processing of processing data using a receiver using a comparing means has been proposed to detect a condition change for the large quantities of data at a short cycle and to attain the addition of a condition time at high accuracy and at high resolution.
Abstract: PURPOSE:To detect a condition change for the large quantities of data at a short cycle and to attain the addition of a condition change time at high accuracy and at high resolution by performing the condition change detection processing of processing data using a receiver CONSTITUTION:To a receiver 2 constituting a plant supervisory equipment, a preceding data buffer 22 to hold the received data of last time, a comparing means 23 to compare the contents of the preceding data buffer 22 with the contents of an input buffer to hold the latest data, and a timer 25 to give the condition change time to the data, whose condition change is detected by the comparing means 23 are added Consequently, by detecting the condition change of the received data, adding the condition change time by means of a receiver 7, and distributing the processing, the processing in a supervising processor 8 can be reduced Thus, even when the large quantities of data are inputted, the condition change of the entire data is detected at the short cycle, and the addition of the condition change time can be attained at the high accuracy and at the high resolution

1 citations


Proceedings ArticleDOI
22 May 1989
TL;DR: A comparative analysis of three image-change-detection algorithms is presented, which are: pairwise subtraction of consecutive images; a two-dimensional least-square adaptive algorithm; and an order-recursive least- square lattice lattice adaptive algorithm.
Abstract: A comparative analysis of three image-change-detection algorithms is presented They are: pairwise subtraction of consecutive images; a two-dimensional least-square adaptive algorithm; and an order-recursive least-square lattice adaptive algorithm These algorithms are representatives of available procedures for image change detection in cluttered environments The analysis is supported by computer simulations, using both recorded and computer-generated images Discussions of real-time processor requirements for these algorithms are also given >

Patent
15 Mar 1989
TL;DR: In this article, a correction coefficient value on the basis of an azimuth change detection value and the map data memory is obtained to perform dynamic correction to an error factor, which can be used to enhance the reliability of a position measuring apparatus.
Abstract: PURPOSE:To enhance the reliability of a position measuring apparatus, by obtaining a correction coefficient value on the basis of an azimuth change detection value and the azimuth change elucidated from a map data memory means to perform dynamic correction to an error factor. CONSTITUTION:A position operating and estimating means 5 operates and estimates the present position on the basis of the signal from the first and second number-of-rotation detection means 3a, 3b. Next, a correction coefficient value operating means 6 inputs the azimuth change detection value from the position operating and estimating means 5 and the azimuth change elucidated from a map data memory means 4 to operate a correction coefficient value. Subsequently, the correction coefficient value from the correction coefficient operation means 6 is inputted to the position operating and estimating means 5 to calculate azimuth change quantity. Next, the calculated azimuth change quantity is outputted to the correction coefficient value operating means 6 and the present position data is displayed on a display means 7 through the position operating and estimating means 5. By this method, a dynamically changing azimuth change detection error factor can be corrected automatically and dynamically and the reliability of the apparatus can be enhanced.

Patent
11 Jan 1989
TL;DR: In this article, a change point detection signal of a reproducing signal from a recording medium is used to improve the detection capability of synchronizing information by detecting the production of change points detection signals of a prescribed number or over in the signals processed in parallel.
Abstract: PURPOSE:To improve the detection capability of synchronizing information by detecting a change point of a reproducing signal from a recording medium, applying parallel processing to the signal and detecting the production of change point detection signals of a prescribed number or over in the signals processed in parallel so as to use the said signal as a synchronization detection signal CONSTITUTION:A signal 6 reproduced from a recording medium and a reproduced clock signal 7 are given to a difference circuit 17, which detects a change point of synchronizing information such as '01' and gives a signal 18 A shift register 18 uses the recovered clock signal 7 to apply parallel processing to the signal 18 A majority decision logic circuit 12 outputs a signal 13 if signals of a prescribed number or over exist in the parallel signals 19 from the shift register 8 Moreover, a synchronizing detection gate signal 14 and an AND gate are used to avoid an error due to the production of a similar pattern in the data to generate a synchronizing detection signal 16 A gate signal is obtained from a detection signal or the like for a sector mark Through the constitution above, the recording detector taking stable data synchronization is obtained even in the recording and reproducing system with high error production probability

Patent
31 Jan 1989
TL;DR: In this paper, the change detection signal is outputted to detect accurately the change timing of a process status while excluding the effect of an external disturbance by using a shift register, a change detection pattern and a change discrimination means or the like.
Abstract: PURPOSE:To prevent fault of installation or defective processing by providing a shift register, a change detection pattern and a change discrimination means or the like so as to detect accurately the change timing of a process status while excluding the effect of an external disturbance. CONSTITUTION:A DI history table 2 updates and stores a signal fetched at a prescribed sampling period by a change point detector 1 into a shift register having a bit number corresponding to a prescribed sampling number as a time series binary pattern signal. Moreover, the correct reference change state of the process status in a prescribed sampling period width before and after the process status change point is preset to an ON-change detection pattern 4 and an OFF-channel detection pattern 5. Then the master pattern 3 extracts only the signal required to collate the pattern 4 or 5 among the input pattern signals stored in the table 2. Then the pattern is collated for each content update of the register via the device 1 and when they are coincident, the change detection signal is outputted to detect accurately the process change timing.

Proceedings ArticleDOI
21 Mar 1989
TL;DR: In this article, an important application of vision in aerial reconnaissance is the detection and classification of changes between images of the same scene taken at different times, and the disparity field must be computed in order for the images to be compared.
Abstract: An important application of vision in aerial reconnaissance is the detection and classification of changes between images of the same scene taken at different times. There are a variety of factors that make this an exceedingly difficult problem: 1) The images are ordinarily taken from (slightly) different vantage points. If the images are taken at sufficiently high altitudes, then the images can, in principle, be registered by a single global coordinate transformation. Otherwise (and this is often the case) the images must be treated as a stereo pair, and the corresponding disparity field must be computed in order for the images to be compared.

Journal ArticleDOI
TL;DR: In this article, the authors considered the abrupt change detection problem for systems of ARMAX type, where the instant of abrupt change occurence is unknown and the detection principle is based on the stochastic aspect of the system.