Optimization of Piecewise Constant Approximation for Segmented Image

Харинов, Михаил Вячеславович; Khanykov, Igor G.

doi:10.15622/sp.40.12

Cited by 13 publications

(9 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The second stage "b" forms N sp superpixels, in fact, performing an improvement in the quality of a given partition with a fixed number of colors (clusters). For this task, two basic algorithms have been developed: SI-method (Segmentation Improvement) [2] and K-meanless method (K-means-without-means method) [3]. Many software implementations are due to both the possibility of a combination of a pair of basic SI and K-meanless methods (separately, sequentially, cyclically), and versions of the methods themselves (segmental, cluster).…”

Section: The Modification Of the Classical Ward's Methodsmentioning

confidence: 99%

“…The combination of "merge&divide" operations ( 1), ( 3) improves image quality in the SI-method by dividing one of the segments (numbered 1) and merging the other two mismatched (numbered 2 and 3). SI-method is executed by the criterion [2]:…”

Section: The Software-algorithmic Toolkitmentioning

confidence: 99%

“…Present paper is devoted to the application of the modification [2,3] of one of the classical cluster analysis methods [4,5] for preprocessing digital images of Earth remote sensing at the segmentation stage. Section 2 provides a brief overview of classical cluster analysis techniques as applied to image processing.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

The Application of the Model of High-Speed Pixel Clustering in Problems of Preprocessing of the Images of the Remote Sensing of the Earth

Khanykov

2020

Proceedings of the 30th International Conference on Computer Graphics and Machine Vision (GraphiCon 2020). Part 2

View full text Add to dashboard Cite

The purpose of the research is to use the modified Ward’s method in high-speed processing of full-HD images of the remote sensing of the Earth. The classical Ward’s method is modified by dividing the computational process into three successive stages. The first stage quickly builds a coarse hierarchy of approximations. The second stage performs a quality improvement of the specified partition for a fixed number of colors (clusters). The third stage is the clustering of the superpixels using the Ward’s method. The software-algorithmic toolkit consists of four operations on clusters of pixels and image segments: merge operation joins together two clusters; divide operation reversibly disjoins the selected cluster into two; split operation extracts the part of the cluster into individual cluster; correct operation reclassifies pixels by extracting from one cluster and inserting into another cluster. The quality is assessed by the total squared error. The quality improvement is provided by iterative execution of a combination of merge and divide operations of pixel clusters, in particular image segments. One of the clusters (segments) is divided in two and a pair of other mismatched with it is combined into one according to the criterion of the minimum increment of the total squared error. The proposed modified Ward’s method is appropriate in processing of fullHD images of the remote sensing of the Earth. The results of processing in pure segmentation and clustering modes are compared. The proposed pixel clustering model is appropriate in high-speed processing of the fullHD images. The pixel clustering in comparison with image segmentation allows to define in more detail both the contours of objects of interest and their internal structure

show abstract

Section: The Modification Of the Classical Ward's Methodsmentioning

confidence: 99%

Section: The Software-algorithmic Toolkitmentioning

confidence: 99%

See 1 more Smart Citation

The Application of the Model of High-Speed Pixel Clustering in Problems of Preprocessing of the Images of the Remote Sensing of the Earth

Khanykov

2020

Proceedings of the 30th International Conference on Computer Graphics and Machine Vision (GraphiCon 2020). Part 2

View full text Add to dashboard Cite

show abstract

“…Currently, there exist various methods and approaches to the solution of image-based segmentation problem [1][2][3]. The most promising solutions in this domain are the approaches to the segmentation problem, which involve neural networks [4][5][6][7][8][9].…”

Section: Related Workmentioning

confidence: 99%

Approach to Image-Based Segmentation of Complex Surfaces Using Machine Learning Tools During Motion of Mobile Robots

Rubtsova

2021

Electromechanics and Robotics

View full text Add to dashboard Cite

In development of formation control systems for modular robotic devices, especially relevant is the problem of analysis and, particularly, segmentation of complex surfaces, down which the robotic device will move. An approach based on fine-tuning of a neural network model HRNet was developed to solve the problem of segmentation of complex surfaces. Model fine-tuning was performed based on a custom dataset, which included 15,000 labeled images. The training dataset included the scenes of the following types: scenes with stairways, scenes with even surfaces, scenes with isolated obstacles and with groups of obstacles. Approbation and functional quality assessment of the developed approach were performed based on the test dataset, which included 3000 images with different levels of scene illumination. According to the results of the testing, the developed approach shows decent quality of segmentation on images with even surfaces (IoU = 90.2%) and with stairways (IoU = 71.3%) as well maintains some resilience to the variations in the scene luminosity levels. After fine-tuning, the averaged performance metrics of this neural network model on images with luminosity levels of 100% and 70% increased by 9.1% and 7.3% at average and resulted in 65.5% and 52%, respectively.

show abstract

“…In the works of Khanykov et al , 2017; Kharinov and Khanykov, 2015a, b; Khanykov, 2017, a digital image segmentation algorithm is proposed, the flowchart of which is presented in Figure 4. The segmentation algorithm generates the hierarchical sequence of the source image segmentations, divided into clusters from 1 to N .…”

Section: Quasi-optimal Segmentation Algorithmmentioning

confidence: 99%

The classification of the image segmentation algorithms

Khanykov

Tolstoj

Levonevskiy

2020

IJIUS

Self Cite

View full text Add to dashboard Cite

PurposeThe purpose of this paper is the image segmentation algorithms (ISA) classification analysis, providing for advanced research and design of new computer vision algorithms.Design/methodology/approachFor the development of the required algorithms a three-stage flowchart is suggested. An algorithm of quasi-optimal segmentation is discussed as a possible implementation of the suggested flowchart. A new attribute is introduced reflecting the specific hierarchical algorithm group, which the proposed algorithm belongs to. The introduced attribute refines the overall classification scheme and the requirements for the algorithms under development.FindingsOptimal approximation generation is a computationally intensive task. The computational complexity can be reduced using a hierarchical data framework and a set of auxiliary algorithms, contributing to overall quality improvement. Because hierarchical solutions usually are distinctively suboptimal, further optimization to them was applied. A new classification attribute, proposed in this paper allows to discover previously hidden «blank spots», having decomposed the two-tier ISA classification scheme. The new classification attribute allows to aggregate algorithms, yielding multiple partitions at output and assign them to a dedicated group.Originality/valueThe originality of the paper consists in development of a high-level ISA classification, as well in introduction of a new classification attribute, pertinent to iterative algorithm groups and to hierarchically structured data presentation algorithms.

show abstract

Optimization of Piecewise Constant Approximation for Segmented Image

Cited by 13 publications

References 0 publications

The Application of the Model of High-Speed Pixel Clustering in Problems of Preprocessing of the Images of the Remote Sensing of the Earth

The Application of the Model of High-Speed Pixel Clustering in Problems of Preprocessing of the Images of the Remote Sensing of the Earth

Approach to Image-Based Segmentation of Complex Surfaces Using Machine Learning Tools During Motion of Mobile Robots

The classification of the image segmentation algorithms

Contact Info

Product

Resources

About