A new method for multi-texture segmentation using neural networks

Martins, Allan de Medeiros; Filho, W.T. de Almeida; Brito, Agostinho M.; Neto, Alexandre

doi:10.1109/ijcnn.2002.1007459

Cited by 3 publications

(3 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such a representation does not retain any information about the shapes of objects in images and obtains moderate results. Other systems [16,26,35,52] rely on texture detection. Texture is characterized by the repetition of basic elements or textons.…”

Section: Context and Related Workmentioning

confidence: 99%

Semantic-enriched visual vocabulary construction in a weakly supervised context

Rizoiu

Velcin

Lallich

2015

IDA

View full text Add to dashboard Cite

One of the prevalent learning tasks involving images is content-based image classification. This is a difficult task especially because the low-level features used to digitally describe images usually capture little information about the semantics of the images. In this paper, we tackle this difficulty by enriching the semantic content of the image representation by using external knowledge. The underlying hypothesis of our work is that creating a more semantically rich representation for images would yield higher machine learning performances, without the need to modify the learning algorithms themselves. The external semantic information is presented under the form of non-positional image labels, therefore positioning our work in a weakly supervised context. Two approaches are proposed: the first one leverages the labels into the visual vocabulary construction algorithm, the result being dedicated visual vocabularies. The second approach adds a filtering phase as a pre-processing of the vocabulary construction. Known positive and known negative sets are constructed and features that are unlikely to be associated with the objects denoted by the labels are filtered. We apply our proposition to the task of content-based image classification and we show that semantically enriching the image representation yields higher classification performances than the baseline representation.

show abstract

Section: Context and Related Workmentioning

confidence: 99%

Semantic-enriched visual vocabulary construction in a weakly supervised context

Rizoiu

Velcin

Lallich

2015

IDA

View full text Add to dashboard Cite

show abstract

“…By tracing back to few decades, several approaches regarding to Auto Brightness Contrast (ABC) techniques had been proposed to enhance the image contrast. Though they achieved some degree of success (Tolat et al ., ; De Medeiros Martins et al ., ; Sim et al ., , ) but the conventional ABC methods still restrict by its constraints such that the region of interest (ROI) does not occupy a significant portion of field‐of‐vision (FOV).…”

Section: Introductionmentioning

confidence: 99%

Local dynamic range compensation for scanning electron microscope imaging system

Sim

Huang

2015

Scanning

View full text Add to dashboard Cite

This is the extended project by introducing the modified dynamic range histogram modification (MDRHM) and is presented in this paper. This technique is used to enhance the scanning electron microscope (SEM) imaging system. By comparing with the conventional histogram modification compensators, this technique utilizes histogram profiling by extending the dynamic range of each tile of an image to the limit of 0-255 range while retains its histogram shape. The proposed technique yields better image compensation compared to conventional methods.

show abstract

“…During the last decade, several approaches based on auto brightness contrast (ABC) compensation techniques were proposed to solve the considered problem and have had different degrees of success (De Medeiros Martins et al 2002, Tolat et al 1991. Though their achievements were considerable, conventional ABC methods still have several constraints when the region-of-interest (ROI) does not occupy a significant portion of field-of-view (FOV).…”

Section: Introductionmentioning

confidence: 99%

A real-time image dynamic range compensation for scanning electron microscope imaging system

2006

View full text Add to dashboard Cite

Summary: This paper presents the development and implementation of a real-time dynamic range compensation system for scanning electron microscope (SEM) imaging applications. Compared with conventional automatic brightness contrast compensators that are based on the average image or pixel intensity level, the proposed system utilizes histogram-profiling techniques to compensate continuously the dynamic range of the processed video signal. The algorithms are implemented in software with a frame grabber card forming the front-end video capture element. The proposed technique yields better image compensation compared with conventional methods.

show abstract

A new method for multi-texture segmentation using neural networks

Cited by 3 publications

References 7 publications

Semantic-enriched visual vocabulary construction in a weakly supervised context

Semantic-enriched visual vocabulary construction in a weakly supervised context

Local dynamic range compensation for scanning electron microscope imaging system

A real-time image dynamic range compensation for scanning electron microscope imaging system

Contact Info

Product

Resources

About