A Document Image Segmentation System Using Analysis of Connected Components

Zirari, Fattah; Ennaji, Abdellatif; Nicolas, Stéphane; Mammass, Driss

doi:10.1109/icdar.2013.154

Cited by 18 publications

(16 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Methods for the segmentation of images of scanned documents are categorized in the scientific literature into block processing methods, pixel processing methods, and methods for processing connected components [3][4][5][6][7][8][9][10][11][12].…”

Section: Literature Review and Problem Statementmentioning

confidence: 99%

“…Known methods for the segmentation of images of scanned documents [2][3][4][5][6][7][8][9][10][11][12] are not capable to simultaneously satisfy modern requirements to the quality of segmentation and to the high-speed performance of segmentation of images of scanned documents. Block processing methods [3,4] are characterized by high-speed performance, but the result of segmentation depends on the partition of the processed image into blocks.…”

Section: Introductionmentioning

confidence: 99%

“…At the same time, pixel processing [5][6][7][8][9] requires a significant time cost although it makes it possible to obtain high quality of segmentation. For methods that involve analysis of connected components [2,[10][11][12], the processing time and the quality of segmentation are determined based on the result of the identification of connected components.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Combined method for scanned documents images segmentation using sequential extraction of regions

Polyakova

Ishchenko

Volkova

et al. 2018

EEJET

View full text Add to dashboard Cite

Запропоновано комбiнований метод сегментацiї зображень вiдсканованих документiв, в якому, на вiдмiну вiд вiдомих, проводиться попереднє вiдокремлення областi графiчних i фотозображень вiд текстових областей i фону. При цьому проводиться аналiз зв'язкових компонент, якi є рiзними для графiчних зображень, фотозображень та текстових областей. Для класифiкацiї видiлених областей, на областi фото i графiки використовується блоковий метод. Встановлено, що такий спосiб розбиття областей на блоки менше впливає на якiсть сегментування в порiвняннi з застосуванням блочного методу безпосередньо до вихiдного зображення. Для вiддiлення бiльш складних за формою текстових областей вiд фону застосовано обробка околицi кожного пiкселя. Для видiлення на зображеннях вiдсканованих документiв границь iлюстрацiй використовувався метод Блумберга. Для подiлу на фото i графiку запропоновано розбиття iлюстрацiї на блоки пiкселiв. Кожному блоку пiкселiв вiдповiдає вектор з двох ознак: середнього значення величини локального градiєнта i середнього значення функцiї, що локалiзує на зображеннях вiдсканованих документiв лiнiйнi об'єкти (графiка i символи тексту). Отриманi вектора ознак класифiкувалися машиною опорних векторiв. При видiленнi текстових фрагментiв використовувалися низькочастотна фiльтрацiя i порогове перетворення. Практичне вiдпрацювання комбiнованого методу проведено для сегментацiї тестових зображень вiдсканованих статей газет з бази даних документiв MediaTeam унiверситету Оулу (Фiнляндiя). Встановлено, що комбiнований метод характеризується пiдвищеною швидкодiєю сегментацiї зображень при високiй якостi обробки Ключовi слова: сегментацiя зображень, вiдсканований документ, блочний метод, графiчне зображення, фотозображеня, текстовий фрагмент, зв'язкова компонента, метод Блумберга UDC 004.93

show abstract

Section: Literature Review and Problem Statementmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Combined method for scanned documents images segmentation using sequential extraction of regions

Polyakova

Ishchenko

Volkova

et al. 2018

EEJET

View full text Add to dashboard Cite

show abstract

“…Connected components analysis based methods have aided image segmentation for the purpose of text detection [16][17][18], brain tissue analysis [19,20] and colour image segmentation using genetic algorithms [21], whereas in our proposed work, we use connected components analysis to refine the boundary map by removing small, isolated connected components (artefacts) which may yield incorrect depth estimates. Our approach may seem similar to [19], where the authors first roughly distinguish the region of non-brain tissue through connected component labelling, and then try to refine those edges using the morphological operations, dilation and erosion.…”

Section: Related Workmentioning

confidence: 99%

Depth-Based Selective Blurring in Stereo Images Using Accelerated Framework

Mukherjee

Guddeti

2014

3D Res

View full text Add to dashboard Cite

We propose a method for depth-based selective blurring of non-interest regions of stereo images using CPU-GPU accelerated framework based on Java Thread Pool and APARAPI.Our method combines both block and region-based stereo matching approaches and generates depth maps using disparity measurements of only 18% image pixels (left or right). The methodology involves segmenting pixel lightness values using fast K-Means implementation, refining the segment boundaries using morphological filtering and connected components analysis; then determining the boundaries' disparities using SAD cost function. Complete disparity maps are reconstructed from these boundaries' disparities. Gaussian blur is applied to de-focus image regions lying in depth ranges of user's non-interest. Experimental results from Middlebury dataset show that our method outperforms traditional disparity estimation approaches using SAD and NCC by up to 33.6% and some recent methods by up to 6.1%. Further, our acceleration framework achieves speed-up of 5.8 for 250 stereo video frames (4096x2304 resolution).

show abstract

“…In this paper, we assume that a neighboring pixel is defined in terms of 8-connectivity since it is more precise in revealing image structures than the other one. Connected components of edge pixels have been widely used in many applications, e.g., image segmentation [15], contour extraction [7], [8], [12], and text detection [4], [13]. However, a connected component of edge pixels may not form a simple discrete curve due to the existence of bifurcation pixels, which limits its application in shape modeling.…”

Section: Introductionmentioning

confidence: 99%

Detection of Bifurcation Pixels in Edge Images

Gong

2014

2014 5th International Conference on Digital Home

View full text Add to dashboard Cite

Connected components of edge pixels are valuable for many applications, e.g., image segmentation, contour extraction, and text detection. However, a connected component of edge pixels may not form a simple curve due to the existence of bifurcation pixels, and thus limits its application in shape modeling. In this paper, we propose an approach to remove bifurcation pixels in an edge image, which let us remove bifurcation pixels in every connected component of edge pixels in an edge image. We present a case study of license plate detection to demonstrate an application of the proposed approach.

show abstract

A Document Image Segmentation System Using Analysis of Connected Components

Cited by 18 publications

References 6 publications

Combined method for scanned documents images segmentation using sequential extraction of regions

Combined method for scanned documents images segmentation using sequential extraction of regions

Depth-Based Selective Blurring in Stereo Images Using Accelerated Framework

Detection of Bifurcation Pixels in Edge Images

Contact Info

Product

Resources

About