Texture Classification Based on Discriminative Features Extracted in the Frequency Domain

Lillo, Antonella Di; Motta, Giovanni; Storer, James A.

doi:10.1109/icip.2007.4379090

Cited by 13 publications

(11 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From these experiments, we observe that our method significantly outperforms those reported in [16,17,28,34], regardless of the selected regularization method, and performs best for all but two of the images. Moreover, while the purely reconstructive framework already provides good results, we observe that the discriminative one noticeably improves the classification rate except for examples 5 and 12.…”

Section: Texture Segmentation Of the Brodatz Datasetmentioning

confidence: 60%

“…Some qualitative results are presented in Figure 4. [16,17,34] and the best results reported in [28]. R1 and R2 denote the reconstructive approach, while D1 and D2 stand for the discriminative one.…”

Section: Texture Segmentation Of the Brodatz Datasetmentioning

confidence: 99%

“…Second, the sparse local representations are learned with an explicit discriminative goal, making the proposed model very different from traditional reconstructive ones [8,27,34]. We illustrate the benefits of this approach in a texture segmentation task on the Brodatz dataset [28] for which it significantly improves over the state of the art [16,17,34], and also present preliminary results showing that it can be used to learn discriminative key patches from the Pascal VOC06 [26] database, and perform the weakly supervised form of feature selection advocated in [25,36] on images from the Graz02 dataset [24]. Section 2 presents the classical framework for learning dictionaries for the sparse representation of overlapping image patches.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Discriminative learned dictionaries for local image analysis

Mairal

Bach

Ponce³

et al. 2008

2008 IEEE Conference on Computer Vision and Pattern Recognition

623

535

View full text Add to dashboard Cite

Sparse signal models have been the focus of much recent research, leading to (or improving upon) state-of-the-art results in signal, image, and video restoration. This article extends this line of research into a novel framework for local image discrimination tasks, proposing an energy formulation with both sparse reconstruction and class discrimination components, jointly optimized during dictionary learning. This approach improves over the state of the art in texture segmentation experiments using the Brodatz database, and it paves the way for a novel scene analysis and recognition framework based on simultaneously learning discriminative and reconstructive dictionaries. Preliminary results in this direction using examples from the Pascal VOC06 and Graz02 datasets are presented as well.

show abstract

Section: Texture Segmentation Of the Brodatz Datasetmentioning

confidence: 60%

Section: Texture Segmentation Of the Brodatz Datasetmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Discriminative learned dictionaries for local image analysis

Mairal

Bach

Ponce³

et al. 2008

2008 IEEE Conference on Computer Vision and Pattern Recognition

623

535

View full text Add to dashboard Cite

show abstract

“…2 A multi-scale version, considering several patch sizes, has also been studied. The authors of [48] have then proposed to extract texture discriminative features in the frequency domain by applying a Fourier transform in polar coordinates, followed by dimensionality reduction via PCA (Principal Components Analysis) or the computation of Fisher coefficients. Centroids are then computed for each class with a vector quantization method.…”

Section: A Pre-processing Of the Datamentioning

confidence: 99%

Learning Discriminative Multilevel Structured Dictionaries for Supervised Image Classification

Mazaheri,

Vural,

Labit

et al. 2018

Preprint

View full text Add to dashboard Cite

Sparse representations using overcomplete dictionaries have proved to be a powerful tool in many signal processing applications such as denoising, super-resolution, inpainting, compression or classification. The sparsity of the representation very much depends on how well the dictionary is adapted to the data at hand. In this paper, we propose a method for learning structured multilevel dictionaries with discriminative constraints to make them well suited for the supervised pixelwise classification of images. A multilevel tree-structured discriminative dictionary is learnt for each class, with a learning objective concerning the reconstruction errors of the image patches around the pixels over each class-representative dictionary. After the initial assignment of the class labels to image pixels based on their sparse representations over the learnt dictionaries, the final classification is achieved by smoothing the label image with a graph cut method and an erosion method. Applied to a common set of texture images, our supervised classification method shows competitive results with the state of the art.

show abstract

“…In past work (e.g., Di Lillo, Motta, and Storer [9]) we have developed FPFT, a translation and rotation invariant texture analysis engine that acquires feature vectors as depicted in Figure 9, and then also developed a classification and segmentation technique which employs vector quantization and a variation of the Kuwahara filter. FPFT was tested by using benchmarks from the Outex database (Ojala, Mäenpää, Pietikäinen, Viertola, Kyllönen, and Huovinen [10]), a reference database containing a large collection of problems composed of synthetic and natural textured images presenting various naturally occurring transformations, such as rotation, scaling, and translation.…”

Section: Texturementioning

confidence: 99%

Compression-Based Tools for Navigation with an Image Database

et al. 2012

Self Cite

View full text Add to dashboard Cite

Abstract:We present tools that can be used within a larger system referred to as a passive assistant. The system receives information from a mobile device, as well as information from an image database such as Google Street View, and employs image processing to provide useful information about a local urban environment to a user who is visually impaired. The first stage acquires and computes accurate location information, the second stage performs texture and color analysis of a scene, and the third stage provides specific object recognition and navigation information. These second and third stages rely on compression-based tools (dimensionality reduction, vector quantization, and coding) that are enhanced by knowledge of (approximate) location of objects.

show abstract

Texture Classification Based on Discriminative Features Extracted in the Frequency Domain

Cited by 13 publications

References 14 publications

Discriminative learned dictionaries for local image analysis

Discriminative learned dictionaries for local image analysis

Learning Discriminative Multilevel Structured Dictionaries for Supervised Image Classification

Compression-Based Tools for Navigation with an Image Database

Contact Info

Product

Resources

About