A cooperative particle swarm optimizer with statistical variable interdependence learning

Sun, Liang; Yoshida, Shigeaki; Cheng, Xiaochun; Liang, Yanchun

doi:10.1016/j.ins.2011.09.033

Cited by 71 publications

(35 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another approach attempts to generate segments based on the provision of a selection of exemplar, template or reference segments [46,52,53,[59][60][61][62]. The geometry and other properties of the provided reference segments are used to drive a search or construction process for quality segmentation.…”

Section: Geographic Object-based Image Analysismentioning

confidence: 99%

“…Visiting each cell in a multidimensional space, or brute force search, is computationally impracticable (due to segmentation and to a much lesser extent other processes) [59]. Metaheuristics are commonly preferred to traverse the parameter set of the given segmentation algorithm [51,[59][60][61][67][68][69][70], owing to various advantages over simpler search methods. Most importantly they are derivative-free population-based methods, less likely to fall into local optima as found in Figure 1.5.…”

Section: Automated Segmentation Algorithm Parameter Tuningmentioning

confidence: 99%

“…The selection of an appropriate method is typically conducted using experimentalism [20]. The utilized optimisers progressed from classic genetic algorithms [21] to proven faster methods such as differential evolution [59] and direct search derivative free methods [60]. For more complex optimization problems the use of metaheuristics are warranted over simpler search strategies [48].…”

Section: Advances In Search Centric Sample Supervised Segment Generationmentioning

confidence: 99%

“…The hybridized image is passed on to a given segmentation algorithm ( Figure 5.3), where the optimizer also search the parameter space of said algorithm, due to parameter domain interdependencies [29,60]. In this work two segmentation algorithms are tested, namely Multiresolution Segmentation (MS) [12] and Simple Linear Iterative Clustering (SLIC) [61].…”

Section: Search Landscape: Segmentation Sub-componentmentioning

confidence: 99%

See 3 more Smart Citations

Sample Supervised Search‐Centric Approaches in Geographic Object‐Based Image Analysis

Fourie

Schöepfer

2017

Earth Observation for Land and Emergency Monitoring

View full text Add to dashboard Cite

Section: Geographic Object-based Image Analysismentioning

confidence: 99%

Section: Automated Segmentation Algorithm Parameter Tuningmentioning

confidence: 99%

Section: Advances In Search Centric Sample Supervised Segment Generationmentioning

confidence: 99%

Section: Search Landscape: Segmentation Sub-componentmentioning

confidence: 99%

See 2 more Smart Citations

Sample Supervised Search‐Centric Approaches in Geographic Object‐Based Image Analysis

Fourie

Schöepfer

2017

Earth Observation for Land and Emergency Monitoring

View full text Add to dashboard Cite

“…The hybridized image is passed on to a given segmentation algorithm (Figure 3), where the optimizer also search the parameter space of said algorithm, due to parameter domain interdependencies [29,60]. In this work two segmentation algorithms are tested, namely Multiresolution Segmentation (MS) [12] and Simple Linear Iterative Clustering (SLIC) [61].…”

Section: Search Landscape: Segmentation Sub-componentmentioning

confidence: 99%

Classifier Directed Data Hybridization for Geographic Sample Supervised Segment Generation

Fourie

Schöepfer

2014

Remote Sensing

View full text Add to dashboard Cite

Quality segment generation is a well-known challenge and research objective within Geographic Object-based Image Analysis (GEOBIA). Although methodological avenues within GEOBIA are diverse, segmentation commonly plays a central role in most approaches, influencing and being influenced by surrounding processes. A general approach using supervised quality measures, specifically user provided reference segments, suggest casting the parameters of a given segmentation algorithm as a multidimensional search problem. In such a sample supervised segment generation approach, spatial metrics observing the user provided reference segments may drive the search process. The search is commonly performed by metaheuristics. A novel sample supervised segment generation approach is presented in this work, where the spectral content of provided reference segments is queried. A one-class classification process using spectral information from inside the provided reference segments is used to generate a probability image, which in turn is employed to direct a hybridization of the original input imagery. Segmentation is performed on such a hybrid image. These processes are adjustable, interdependent and form a part of the search problem. Results are presented detailing the performances of four method variants compared to the generic sample supervised segment generation approach, under various conditions in terms of resultant segment quality, required computing time and search process characteristics. Multiple metrics, metaheuristics and segmentation algorithms are tested with this approach. Using the spectral data contained within user OPEN ACCESS Remote Sens. 2014, 6 11853 provided reference segments to tailor the output generally improves the results in the investigated problem contexts, but at the expense of additional required computing time.

show abstract

A comparison of swarm intelligence algorithms for structural engineering optimization

Parpinelli

Teodoro

Lopes

2012

Numerical Meth Engineering

View full text Add to dashboard Cite

SUMMARY This paper compares the performance of three swarm intelligence algorithms for the optimization of hard engineering problems. The algorithms tested were bacterial foraging optimization (BFO), particle swarm optimization (PSO), and artificial bee colony (ABC). Besides the regular BFO, two other variants reported in the literature were also included in the study: adaptive BFO and swarming BFO. Both PSO and ABC were tested using the regular algorithm and variants that include explosion (mass extinction). Three optimization problems of structural engineering were used: minimization of the cost of a welded beam, minimization of the construction cost of a pressure vessel, and minimization of the total weight of a 10‐bar plane truss. All problems are strongly constrained. The algorithms were evaluated using two criteria: quality of solutions and the number of function evaluations. The results show that PSO presented the best balance between these two criteria. For the optimization problems approached in this paper, we can also conclude that the explosion procedure resulted in no significant improvements. Copyright © 2012 John Wiley & Sons, Ltd.

show abstract

A cooperative particle swarm optimizer with statistical variable interdependence learning

Cited by 71 publications

References 20 publications

Sample Supervised Search‐Centric Approaches in Geographic Object‐Based Image Analysis

Sample Supervised Search‐Centric Approaches in Geographic Object‐Based Image Analysis

Classifier Directed Data Hybridization for Geographic Sample Supervised Segment Generation

A comparison of swarm intelligence algorithms for structural engineering optimization

Contact Info

Product

Resources

About