Soft Computing and Image Analysis: Features, Relevance and Hybridization

Pal, Shantanu; Ghosh, Ashish; Kundu, Malay K.

doi:10.1007/978-3-7908-1858-1_1

Cited by 6 publications

(8 citation statements)

References 75 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Image analysis and pattern recognition are fields in which such ambiguity is inherent [1]. This ambiguity is a function of the variations in color intensity, hue, and noise in photographic images.…”

Section: Introductionmentioning

confidence: 98%

Soft computing algorithms applied to the segmentation of nerve cell images

Schäfer

Hammell

2005

Sixth International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing

View full text Add to dashboard Cite

Microscopic images of stained nerve cells are routinely analyzed during neuropathological research. Manual analysis relies heavily on operator knowledge, and therefore can be highly subjective. The process is also time consuming. This paper investigates the use of Fuzzy C-Means to automate the analysis of nerve cell images. Using Fuzzy C-Means clustering, nerve cells are detected in an image. The nerve cells are then classified into degrees of health based upon their physical characteristics. A fuzzy approach is taken in order to account for vagueness in the data. This ambiguity stems from both the nature of digital images and the nature of biological systems.

show abstract

Section: Introductionmentioning

confidence: 98%

Soft computing algorithms applied to the segmentation of nerve cell images

Schäfer

Hammell

2005

Sixth International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing

View full text Add to dashboard Cite

show abstract

“…The current trend in image reconstruction and recognition systems involves supplementing and partially replacing the classical methods with artificial intelligence systems, which incorporate machine learning algorithms. Various papers [1][2][3][4] have reviewed machine learning methods in image recognition, with particular emphasis on medical applications. It is worth noting that modern image reconstruction and recognition systems typically rely on optimization algorithms with constraints while utilizing appropriately selected regularization methods [5].…”

Section: Introductionmentioning

confidence: 99%

Image Recognition and Reconstruction With Machine Learning: An Inverse Problem Approach

Citko,

Sienko

2023

IEEE Access

View full text Add to dashboard Cite

Image recognition and reconstruction are common problems in many image processing systems. These problems can be formulated as a solution to the linear inverse problem. This article presents a machine learning system model that can be used in the reconstruction and recognition of vectorized images. The analyzed inverse problem is given by the equations 𝐹(𝒙 𝒊 ) = 𝒚 𝑖 and 𝒙 𝑖 = 𝐹 −1 (𝒚 𝑖 ), 𝑖 = 1, … , 𝑁, where 𝐹(•) is a linear mapping for 𝒙 𝑖 ∈ 𝑋 ⊂ 𝑅 𝑛 , 𝒚 𝑖 ∈ 𝑌 ⊂ 𝑅 𝑚 . Thus, 𝒚 𝑖 can be seen as a projection of image 𝒙 𝑖 , and 𝒙 𝑖 should be reconstructed as a solution to the inverse problem. We consider image reconstruction as an inverse problem using two different schemes. The first one, when 𝒙 𝑖 = 𝐹 −1 (𝒚 𝑖 ), can be seen as an operation with associative memory, and the second one, when 𝒙 𝑖 = 𝐹 −1 (𝒚 𝑖 ), can be implemented by creating random vectors for training sets. Moreover, we point out that the solution to the inverse problem can be generalized to complex-valued images 𝒙 𝑖 and 𝒚 𝑖 . In this paper, we propose a machine learning model based on a spectral processor as an alternative solution to deep learning based on optimization procedures.INDEX TERMS Machine learning systems, image reconstruction and recognition, inverse problem.

show abstract

“…Note that the -band wavelet transform is a tool for viewing signals at different scales and decomposes a signal by projecting it onto a family of functions generated from a single wavelet basis via its dilations and translations [13]. Neurofuzzy computing [14], which integrates the merits of fuzzy set theory and artificial neural networks (ANNs), enables the feature selection process artificially more intelligent. Incorporation of fuzzy set theory helps one to deal with uncertainties in remotely sensed images in an efficient manner.…”

Section: Introductionmentioning

confidence: 99%

Segmentation of remotely sensed images using wavelet features and their evaluation in soft computing framework

Acharyya

Kundu

2003

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

Abstract-The present paper describes a feature extraction method based on -band wavelet packet frames for segmenting remotely sensed images. These wavelet features are then evaluated and selected using an efficient neurofuzzy algorithm. Both the feature extraction and neurofuzzy feature evaluation methods are unsupervised, and they do not require the knowledge of the number and distribution of classes corresponding to various land covers in remotely sensed images. The effectiveness of the methodology is demonstrated on two four-band Indian Remote Sensing 1A satellite (IRS-1A) images containing five to six overlapping classes and a three-band SPOT image containing seven overlapping classes.Index Terms-Adaptive basis selection, fuzzy feature evaluation index, -band wavelet packet frames, neural networks, remotely sensed image.

show abstract

Soft Computing and Image Analysis: Features, Relevance and Hybridization

Cited by 6 publications

References 75 publications

Soft computing algorithms applied to the segmentation of nerve cell images

Soft computing algorithms applied to the segmentation of nerve cell images

Image Recognition and Reconstruction With Machine Learning: An Inverse Problem Approach

Segmentation of remotely sensed images using wavelet features and their evaluation in soft computing framework

Contact Info

Product

Resources

About