# T. Pavlidis, "Waveform Segmentation Through Functional Approximation, " IEEE Trans. Computers, vol. 22, no. 7, p. 689, 1973.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....would improve dimension reduction performance. Here, we introduce dynamic segmentation. Dynamic segmentation refers to partitioning feature trajectories dynamically so as to minimize the average dimensionality. However, finding optimal partitioning requires a huge amount of calculation (e.g. [4]) Thus, our technique addresses quick suboptimal partitioning of the trajectories by modifying the formulation of dynamic segmentation and using the discreteness of dimension changes. It also achieves theoretical optimality in the amount of calculation to derive the suboptimal partitioning under ....

T. Pavlidis: "Waveform segmentation through functional approximation", IEEE Trans. on Computers, Vol. C-22, No. 7, pp. 689-697, 1973.

....C of a time series C is optimal (in terms of the quality of approximation) iff C has the least reconstruction error among all possible Msegment APCA representations of C. Finding the optimal piecewise polynomial representation of a time series requires a O(Mn 2) dynamic programming algorithm [15, 35]. This is too slow for high dimensional data. In this paper, we propose a new algorithm to produce almost optimal APCA representations in O(nlog(n) time. The algorithm works by first converting the problem into a wavelet compression problem, for which there are well known optimal solutions, then ....

....2 and 3 is the best pair to merge as it results in the minimum increase in reconstruction error. Figure 6(d) shows the final 3 segment APCA representation of C produced by the Compute APCA algorithm. We experimentally compared this algorithm with several of the heuristic, merging algorithms [15, 35, 42] and found it is faster (at least 5 times faster for any length time series) and slightly superior in terms of reconstruction error. 3.3 Distance measures defined for the APCA representation Suppose we have a time series C, which we convert to the APCA representation C, and a query time series ....

Pavlidis, T. (1976). Waveform segmentation through functional approximation. IEEE Transcations on Computers, Vol C-22, NO. 7 July.

....C) Space Shuttle telemetry. D) Electrocardiogram. E) Manufacturing. F) Exchange rate. Figure 5: A time series C and its APCA representation C, with M = 4 In general, finding the optimal piecewise polynomial representation of a time series requires a O(Nn 2 ) dynamic programming algorithm [15, 35]. For most purposes, however, an optimal representation is not required. Most researchers, therefore, use a greedy suboptimal approach instead [42, 27, 46] In this work we utilize an original algorithm which produces high quality approximations in O(nlog(n) The algorithm works by first ....

Pavlidis, T. (1976). Waveform segmentation through functional approximation. IEEE Transcations on Computers, Vol C-22, NO. 7 July.

....i th segment can be calculated as cr i cr i 1 . Figure 5 illustrates this notation. Figure 5: A time series C and its APCA representation C, with M = 4 In general, finding the optimal piecewise polynomial representation of a time series requires a O(Nn 2 ) dynamic programming algorithm [15, 35]. For most purposes, however, an optimal representation is not required. Most researchers, therefore, use a greedy suboptimal instead [42, 27, 46] In this work we utilize an original algorithm which produces high quality approximations in O(nlog(n) The algorithm works by first converting the ....

Pavlidis, T. (1976). Waveform segmentation through functional approximation. IEEE Transcations on Computers, Vol C-22, NO. 7 July.

....developed for detecting breakpoints. Breakpoint detection algorithms can be roughly grouped into two categories: one is based on the detection directly from the underlying images [3] 6] 19] 14] the other is based on digital arcs, resulting from edge detection and linking [2] 11] 8] 1] [9], 10] 17] 18] The research described in this paper is concerned with detecting breakpoints from digital arcs. Breakpoint detection from digital arcs partitions a given digital arc sequence into digital arc subsequences having the property that each arc subsequence is a maximal length sequence ....

....from digital arc sequences. The basis for these techniques is to identify the locations of the endpoints of each maximal line segment. Different criteria have been proposed for detecting breakpoints including maximum curvature, deflection angle, maximum deviation, and total fitting errors [1] [9], 10] 17] 8] 2] 11] 18] A major problem with the existing approaches is that the employed criterion is not tied to a statistical analysis involving an explicit noise model, therefore making the method statistically inefficient with regard to noise. To overcome this, we have developed a ....

[Article contains additional citation context not shown here]

T. Pavlidis. Waveform segmentation through functional approximation. IEEE Tran. on Computers, 22(7):689, 1973.

.... an intuitive and practical method for representing curves in a simple parametric form (generalizations to low order polynomial and spline representations are straightforward) There are a large number of different algorithms for segmenting a curve into the K best piecewise linear segments (e.g. Pavlidis (1974)) We use a computationally efficient and flexible approach based on bottom up merging of local segments into a hierarchical multi scale 0 0.5 1 1.5 2 2.5 3 x 10 4 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Seconds Figure 1: Automated segmentation of an inertial navigation sensor from Space ....

Pavlidis, T., `Waveform segmentation through functional approximation,' IEEE Trans. Comp., C-22, no.7, 689--697, 1974.

....from digital arc sequences. The basis for these techniques is to identify the locations of the endpoints of each maximal line segment. Different criteria have been proposed for detecting corner points including maximum curvature, deflection angle, maximum deviation, and total fitting errors [1][5][6] 11] A major problem with existing approaches is that the employed criterion is not tied to a statistical analysis, therefore rendering existing methods susceptible to noise. To overcome this, we present a statistical approach for corner detection. Here, the corner criterion is treated as a ....

....optimal because not all the points on the arc are used to detect the corner points. This may result in high locational errors. To reduce the location errors with the detected corners, we perform a corner optimization. The corner optimization, based on Pavlidis s discrete optimization method [5], iteratively shifts the detected corner points to produce a better approximation of the arc sequence. While the iterative optimization procedure is guaranteed to terminate with improved location errors, it however may terminate at a local minimum rather than at the global minimum. 4. Performance ....

T. Pavlidis. Waveform segmentation through functional approximation. IEEE Tran. on Computers, 22(7):689, 1973.

No context found.

# T. Pavlidis, "Waveform Segmentation Through Functional Approximation, " IEEE Trans. Computers, vol. 22, no. 7, p. 689, 1973.

No context found.

T. Pavlidis, Waveform Segmentation Through Functional Approximation, IEEE Transactions on Computers C-22 (1973), no. 7, 689--697.

No context found.

Pavlidis, T. (1976). Waveform segmentation through functional approximation. IEEE Transactions on Computers.

No context found.

Pavlidis, T. (1976). Waveform segmentation through functional approximation. IEEE Transactions on Computers.

No context found.

Pavlidis, T., `Waveform Segmentation Through Functional Approximation', IEEE Transactions on Computers, Vol, C-22, NO. 7 July 1976.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC