Automatic image segmentation based on PCNN with adaptive threshold time constant

Wei, Shuo; Qu, Hong; Hou, Mengshu

doi:10.1016/j.neucom.2011.01.005

Cited by 81 publications

(27 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, the neurons with similar intensity always fire synchronously. In view of these advantages, PCNN has been widely applied to image segmentation [25][26][27][28]. In this paper, we segment ROI into a binary image via the SPCNN (simplified PCNN) whose working principle is described in "Appendix."…”

Section: Roi Segmentation By Spcnnmentioning

confidence: 99%

“…In SPCNN, the external input is the normalized image S (obtained by Min-Max normalization), and the series of pulse outputs are the different binary images from which we can pick out the desired segmentation result. For LV segmentation, we set the SPCNN parameters (W, β, V E ) according to [29]: W = [0.5, 1, 0.5; 1, 0, 1; 0.5, 1, 0.5], β = 0.1, V E = 20, and set α e adaptively by using α e = C/mean(S) [28]. C is a constant, and "mean(S)" represents the calculation of the average gray level.…”

Section: Roi Segmentation By Spcnnmentioning

confidence: 99%

“…9, the single neuron structure of SPCNN model includes three parts: input field, modulation field and pulse generator field [28]. And, the discrete SPCNN was described as follows: Fig.…”

Section: Appendixmentioning

confidence: 99%

See 2 more Smart Citations

An SPCNN-GVF-based approach for the automatic segmentation of left ventricle in cardiac cine MR images

Wang

et al. 2016

Int J CARS

View full text Add to dashboard Cite

show abstract

Section: Roi Segmentation By Spcnnmentioning

confidence: 99%

Section: Roi Segmentation By Spcnnmentioning

confidence: 99%

See 1 more Smart Citation

An SPCNN-GVF-based approach for the automatic segmentation of left ventricle in cardiac cine MR images

Wang

et al. 2016

Int J CARS

View full text Add to dashboard Cite

show abstract

“…A PCNN is a novel network that is used in many image processing applications, such as noise removal for gray [17] and color [18] images, image segmentation [19][20][21][22], image smoothing [23], and image enhancement [20,24]. In addition it was successfully explored in many applications such as car plate number detection [25], face recognition [26], and object detection [27].…”

Section: Pulse-coupled Neural Networkmentioning

confidence: 99%

Geometric-attributes-based segmentation of cortical bone slides using optimized neural networks

Hage

Hamade

2015

J Bone Miner Metab

View full text Add to dashboard Cite

In cortical bone, solid (lamellar and interstitial) matrix occupies space left over by porous microfeatures such as Haversian canals, lacunae, and canaliculi-containing clusters. In this work, pulse-coupled neural networks (PCNN) were used to automatically distinguish the microfeatures present in histology slides of cortical bone. The networks' parameters were optimized using particle swarm optimization (PSO). When forming the fitness functions for the PSO, we considered the microfeatures' geometric attributes-namely, their size (based on measures of elliptical perimeter or area), shape (based on measures of compactness or the ratio of minor axis length to major axis length), and a two-way combination of these two geometric attributes. This hybrid PCNN-PSO method was further enhanced for pulse evaluation by combination with yet another method, adaptive threshold (AT), where the PCNN algorithm is repeated until the best threshold is found corresponding to the maximum variance between two segmented regions. Together, this framework of using PCNN-PSO-AT constitutes, we believe, a novel framework in biomedical imaging. Using this framework and extracting microfeatures from only one training image, we successfully extracted microfeatures from other test images. The high fidelity of all resultant segments was established using quantitative metrics such as precision, specificity, and Dice indices.

show abstract

“…So far, thousands of segmentation algorithms have been developed, for which numerous classification methods have been proposed. The traditional image segmentation methods can be classified into the following four classes: (1) threshold-based method [1,2]; (2) edge-based method [3,4]; (3) regionbased method [5,6]; and (4) specific theory-based method [7]. With the development of the artificial neural network, such as fuzzy set theory and graph theory, some novel segmentation algorithms have been proposed in combination with these theories [8][9][10].…”

Section: Introductionmentioning

confidence: 99%

Medical Image Segmentation Using Improved Chan-Vese Model

Guo

Shi

Wang

et al. 2016

Lecture Notes in Electrical Engineering

View full text Add to dashboard Cite

Image segmentation plays a crucial role in many medical imaging applications. The use of the conventional Chan-Vese algorithm for medical image analysis is widespread because of its ability to produce a complete division of the image. However, its drawbacks include over-segmentation and sensitivity to false edges. In this paper, we present a novel algorithm that incorporates k-means clustering and improved Chan-Vese algorithm. The improved Chan-Vese algorithm is based on the similarity between each point and center point in the neighborhood. This algorithm can capture the details of the local region to realize the image segmentation in the gray-level heterogeneous area. The experimental results show that this method can segment the medical image with high accuracy, adaptability and more stable performance compared to the traditional Chan-Vese model.

show abstract

Automatic image segmentation based on PCNN with adaptive threshold time constant

Cited by 81 publications

References 16 publications

An SPCNN-GVF-based approach for the automatic segmentation of left ventricle in cardiac cine MR images

An SPCNN-GVF-based approach for the automatic segmentation of left ventricle in cardiac cine MR images

Geometric-attributes-based segmentation of cortical bone slides using optimized neural networks

Medical Image Segmentation Using Improved Chan-Vese Model

Contact Info

Product

Resources

About