A local fuzzy thresholding methodology for multiregion image segmentation

Aja‐Fernández, Santiago; Curiale, Ariel Hernán; Vegas-Sánchez-Ferrero, Gonzalo

doi:10.1016/j.knosys.2015.02.029

Cited by 132 publications

(100 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…There are 11 built in membership functions available in the Fuzzy logic toolbox. The definition of the membership function can be done by from the histogram of the image or from the centroids, leaving aside histogram information [44]. To obtain a better solution Pseudo Trapezoid-Shaped (PTS) [45] membership functions is utilized.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Implementation of Fuzzy Thresholding for Segmentation of Images

Kumari¹,

Enireddy²

2017

IJCA

View full text Add to dashboard Cite

Segmentation is a process of dividing an image into multiple regions making it easier for the analysis. It is also an important step in Image processing. It also has number of applications which includes Medical field to analyze a disease, scientific fields including, engineering and technology, face recognition and object. Many algorithms and techniques have been developed for the image segmentation. To remove distinctive areas from a picture an immediate and straight forward method is using thresholding. It looks for a worldwide esteem that boosts the partition between yield classes. Noisy or uneven illuminated images pose a challenge for segmentation. The utilization of solitary hard limit esteem is absolutely the wellspring of imperative division mistakes in numerous situations like boisterous images or uneven illumination. In this paper a multiregion thresholding technique is displayed to overcome the normal downsides of thresholding strategies when images are debased with artifacts and commotion. Pixels from the images are related to various yield centroids by means of fuzzy membership function, evading any underlying hard choice. To make this strategy robust to noise and artifacts it utilizes spatial data through a local aggregation step where the participation level of every pixel is adjusted by neighborhood data that considers the enrollments of the encompassing pixels. The results are then compared with the existing techniques and results obtained are satisfactory.

show abstract

Section: Methodsmentioning

confidence: 99%

“…Defuzzification can be used to obtained the output but a maximum operator [44] = { )} is used for obtaining final image.…”

Section: Methodsmentioning

confidence: 99%

Implementation of Fuzzy Thresholding for Segmentation of Images

Kumari¹,

Enireddy²

2017

IJCA

View full text Add to dashboard Cite

show abstract

“…Dalam segmentasi citra, thresholding merupakan salah satu metode yang mudah dan sederhana untuk diimplementasikan [15]. Thresholding merupakan metode yang efektif pada citra menunjukan pemisah yang jelas antara dua cluster pada histogram sehingga proses pemisahan objek dan background mudah dilakukan.…”

Section: Pendahuluanunclassified

Lokal Fuzzy Thresholding Berdasarkan Pengukuran Fuzzy Similarity Pada Interaktif Segmentasi Citra Panoramik Gigi

Gunawan

Arifin

2017

INFOTEL

View full text Add to dashboard Cite

Abstrak -Dalam segmentasi citra, thresholding merupakan salah metode yang mudah dan sederhana untuk diimplementasikan. Pada citra panoramik gigi, penentuan global threshold masih kurang begitu optimal untuk diimplementasikan, hal tersebut dikarenakan adanya faktor penghambat seperti pencahayaan yang tidak merata dan citra yang kabur. Faktor-faktor tersebut dapat menyebabkan histogram tidak bisa dipartisi dengan baik, sehingga akan berpengaruh pada hasil segmentasi. Pada penelitian ini diusulkan local fuzzy thresholding berdasarkan pengukuran fuzzy similarity pada interaktif segmentasi citra panoramik gigi. Metode yang diusulkan terdiri dari tiga tahapan utama, tahap pertama adalah region splitting untuk mendapatkan lokal region. Tahap kedua adalah user marking untuk mendapat initial seed background dan objek, tahap terakhir adalah pengukuran fuzzy similarity pada setiap lokal region untuk mendapatkan nilai lokal threshold. Hasil uji coba pada citra panoramik gigi, metode yang diusulkan berhasil melakukan segmentasi dengan rata-rata missclasification error (ME) 5.47%.Kata kunci -interaktif segmentasi citra, citra panoramik gigi, lokal thresholding, pengukuran fuzzy similarity Abstract -In image segmentation, thresholding is one of the method that is easy and simple to be implemented. In the dental panoramic image, determination of global threshold is still less optimal to be implemented. This is due to inhibiting factors such as uneven contrast and blurred images. These factors can lead to the problem in partitioning histogram, that would-be effect in the segmentation result. In this study, we proposed local fuzzy thresholding based on fuzzy similarity measurement in interactive image segmentation for dental panoramic image. The proposed method consists of three main stages. The first stage is region splitting. This stage is performed to get the local regions. The second stage is the user marking that used to receive initial seed background and object. The last stage is a fuzzy similarity measurement at each local region to get the local threshold value. The results on dental panoramic image show that the proposed method successfully segmenting the images with an average misclassification error (ME) 5.47%.

show abstract

“…In order to assist in this process, various semi-automatic techniques are prevalent for segmenting the object of interest, including methods based on a global threshold [2]; however, a global threshold selection is not straightforward. Use of a single hard threshold is considered a source of segmentation errors [3]. Additionally, the pixels assigned to a single class need not necessarily form a coherent region since spatial locations are ignored.…”

Section: Introductionmentioning

confidence: 99%

“…The result benchmarked with those of deterministic methods like Harris and SUSAN is reported. A segmentation approach using fuzzy threshold for multi-region is formulated in [3] which uses a mapping between centroid of clustered pixels with a membership function. Madasu et al, developed a fuzzy edge and corner detector in case of color images using SUSAN corner detector as a base [29].…”

Section: Introductionmentioning

confidence: 99%