An edge detection by using self-organization

Nagai, Hideaki; Miyanaga, Yoshikazu; Tochinai, Koji

doi:10.1109/icassp.1998.678092

Cited by 7 publications

(4 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…1. edge detectors based on cellular NNs [36], [37], [38], [39], [40], [41], [42], [43]; 2. edge detectors based on self-organizing maps [44], [45], [46]; 3. edge detectors based on Hopfield networks [47], [48], [49], [50]; 4. edge detectors based on multilayer NNs [51], [52], [53], [54]. The above edge detectors, except class 4, are unsupervised ones and, thus, they do not have the function of enhancing the desired edges.…”

Section: Introductionmentioning

confidence: 99%

Neural edge enhancer for supervised edge enhancement from noisy images

Suzuki

Horiba

Sugie

2003

IEEE Trans. Pattern Anal. Machine Intell.

140

View full text Add to dashboard Cite

We propose a new edge enhancer based on a modified multilayer neural network, which is called a neural edge enhancer (NEE), for enhancing the desired edges clearly from noisy images. The NEE is a supervised edge enhancer: Through training with a set of input noisy images and teaching edges, the NEE acquires the function of a desired edge enhancer. The input images are synthesized from noiseless images by addition of noise. The teaching edges are made from the noiseless images by performing the desired edge enhancer. To investigate the performance, we carried out experiments to enhance edges from noisy artificial and natural images. By comparison with conventional edge enhancers, the following was demonstrated: The NEE was robust against noise, was able to enhance continuous edges from noisy images, and was superior to the conventional edge enhancers in similarity to the desired edges. To gain insight into the nonlinear kernel of the NEE, we performed analyses on the trained NEE. The results suggested that the trained NEE acquired directional gradient operators with smoothing. Furthermore, we propose a method for edge localization for the NEE. We compared the NEE, together with the proposed edge localization method, with a leading edge detector. The NEE was proven to be useful for enhancing edges from noisy images.

show abstract

Section: Introductionmentioning

confidence: 99%

Neural edge enhancer for supervised edge enhancement from noisy images

Suzuki

Horiba

Sugie

2003

IEEE Trans. Pattern Anal. Machine Intell.

140

View full text Add to dashboard Cite

show abstract

“…They can be classied into four broad classes: 1) edge detectors based on the cellular NNs [15]{ [18]; 2) those based on the multilayer NN [19,20]; 3) those based on the selforganizing maps [21,22]; 4) those based on the Hopeld networks [23,24]. Most of the edge detectors are unsupervised ones, so they can not necessarily satisfy the requirement (2).…”

Section: 4mentioning

confidence: 99%

Neural Edge Detector - A Good Mimic of Conventional One Yet Robuster against Noise

Suzuki

Horiba

Sugie

2001

Bio-Inspired Applications of Connectionism

View full text Add to dashboard Cite

Abstract. This paper describes a new edge detector using a multilayer neural network, called a neural edge detector (NED), and its capacity for edge detection against noise. The NED is a supervised edge detector: the NED acquires the function of a desired edge detector through training. The experiments to acquire the functions of the conventional edge detectors were performed. The experimental results have demonstrated that the NED is a good mimic for the conventional edge detectors, moreover robuster against noise: the NED can detect the similar edges to those detected by the conventional edge detector; the NED is robuster against noise than the original one is.

show abstract

“…For the details of the clustering procedure, see Refs. [13][14][15][16]. Weights may be introduced, but the clustering is performed in this paper by setting the weights as 1, since one of the goals of the proposed method is to limit the use of parameters.…”

Section: Self-organization Of Clustersmentioning

confidence: 99%

A study of image segmentation based on a robust data clustering method

Ryu

Miyanaga

2004

Electron Comm Jpn Pt III

View full text Add to dashboard Cite

SUMMARYThis paper considers intelligent image clustering and proposes a clustering method based on the self-organizing mechanism, which is an adaptive and autonomous algorithm. The proposed method solves the problem of initial value setting, which has been a difficulty in the conventional clustering method, and has the feature that the effective clusters are always self-organized. A method is realized which is robust and does not require empirical setting of the initial value. Thus, the effective initial cluster can always be selected for various kinds of data sets, and an excellent clustering is obtained. A method is proposed in which these features are utilized in image segmentation.

show abstract

An edge detection by using self-organization

Cited by 7 publications

References 2 publications

Neural edge enhancer for supervised edge enhancement from noisy images

Neural edge enhancer for supervised edge enhancement from noisy images

Neural Edge Detector - A Good Mimic of Conventional One Yet Robuster against Noise

A study of image segmentation based on a robust data clustering method

Contact Info

Product

Resources

About