Sift Descriptors Modeling and Application in Texture Image Classification

Zeglazi, Oussama; Amine, Aouatif; Rziza, Mohammed

doi:10.1109/cgiv.2016.58

Cited by 13 publications

(22 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some popular low-level features are Scale-Invariant Feature Transform (SIFT) [2], Generalized Search Trees (GIST) [3], [4], Histogram of Gradient (HOG) [5], CENsus TRansform hISTogram (CENTRIST) [6], multichannel (mCENTRIST) [7] and OTC [9]. Since these features [2]- [7], [9], [10] exploit the local information on the image, they provide neither the global structural information nor the object information. They may not work properly if the image feature extraction needs global structural details of the image.…”

Section: Related Workmentioning

confidence: 99%

“…For example, if we take an object o i , then its relationship can be shown with lower indexed object o i−1 and higher indexed object o i+1 . The general structure of the objects pattern dictionary (C i ) which shows the semantically related objects for the particular category is shown in (2).…”

Section: A Objects Pattern Dictionarymentioning

confidence: 99%

“…Low-level features [2]- [10] are typically extracted using pixels, colour intensity or texture of the image. These features lack the spatial information on the image, thereby deteriorating the classification accuracy, especially for scene images (indoor/outdoor images).…”

Section: Introductionmentioning

confidence: 99%

“…Domain-specific features are important to solve the specific classification problem. As an example, most of the low-level features, which are based on color or intensity as their features, have poor performance for indoor images involving multiple objects [2].…”

Section: Introductionmentioning

confidence: 99%

“…For instance, two indoor categories library and kitchen may contain similar table and desk objects, but how can we differentiate the images? Similarly, we can hardly find the similarity using traditional features such as [2] if two images of the library category have different structures and the same types of objects (e.g., Fig. 1).…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Indoor Image Representation by High-Level Semantic Features

et al. 2019

View full text Add to dashboard Cite

Indoor image features extraction is a fundamental problem in multiple fields such as image processing, pattern recognition, robotics and so on. Nevertheless, most of the existing feature extraction methods, which extract features based on pixels, color, shape/object parts or objects on images, suffer from limited capabilities in describing semantic information (e.g., object association). These techniques therefore involve undesired classification performance. To tackle this issue, we propose the notion of high-level semantic features and design four steps to extract them. Specifically, we first construct the objects pattern dictionary through extracting raw objects in the images, and then retrieve and extract semantic objects from the objects pattern dictionary. We finally extract our high-level semantic features based on the calculated probability and delta parameter. Experiments on three publicly available datasets (MIT-67, Scene15 and NYU V1) show that our feature extraction approach outperforms state-of-the-art feature extraction methods for indoor image classification, given a lower dimension of our features than those methods.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: A Objects Pattern Dictionarymentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Indoor Image Representation by High-Level Semantic Features

et al. 2019

View full text Add to dashboard Cite

show abstract

Tag-Based Semantic Features for Scene Image Classification

Sitaula

Xiang

Basnet³

et al. 2019

Neural Information Processing

View full text Add to dashboard Cite

The existing image feature extraction methods are primarily based on the content and structure information of images, and rarely consider the contextual semantic information. Regarding some types of images such as scenes and objects, the annotations and descriptions of them available on the web may provide reliable contextual semantic information for feature extraction. In this paper, we introduce novel semantic features of an image based on the annotations and descriptions of its similar images available on the web. Specifically, we propose a new method which consists of two consecutive steps to extract our semantic features. For each image in the training set, we initially search the top k most similar images from the internet and extract their annotations/descriptions (e.g., tags or keywords). The annotation information is employed to design a filter bank for each image category and generate filter words (codebook). Finally, each image is represented by the histogram of the occurrences of filter words in all categories. We evaluate the performance of the proposed features in scene image classification on three commonly-used scene image datasets (i.e., MIT-67, Scene15 and Event8). Our method typically produces a lower feature dimension than existing feature extraction methods. Experimental results show that the proposed features generate better classification accuracies than vision based and tag based features, and comparable results to deep learning based features.Keywords: Image features· semantic features· tags· semantic similarity· tag-based features· search engine.

show abstract