Local Detectors and Descriptors for Object Class Recognition

Khalifa, Faten A.; Semary, Noura A.; El-Sayed, Hatem M.; Hadhoud, Mohiy M.

doi:10.5815/ijisa.2015.10.02

Cited by 6 publications

(4 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To the resulting dataset we also added images distorted with Gaussian blur (σ = 2, σ = 4). Similar transformations can be found in [10]. Finally, 200,000 images have been obtained.…”

Section: Datasetsmentioning

confidence: 75%

See 1 more Smart Citation

Evaluation of Interest Point Detectors in Presence of Noise

Ziomek¹,

Oszust

2016

IJISA

View full text Add to dashboard Cite

Abstract-Detection of repeatable keypoints is often one of the first steps leading to obtain a solution able to recognise objects on images. Such objects are characterised by content of image patches indicated by keypoints. A given image patch is worth being described and processed in further steps, if the interest point inside of it can be found despite different image transformations or distortions. Therefore, it is important to compare keypoint detection techniques using image datasets that contain transformed or noisy images. Since most of detector evaluations rely on small datasets or are focused on a specific application of compared techniques, in this paper two large datasets which cover typical transformations, as well as challenging distortions that can occur while image processing, are used. The first dataset contains 200,000 transformed images, and it has been prepared for the purpose of this study. The second dataset, TID2013, is widely used for perceptual image quality assessment; it contains 3,000 images with 24 distortions. Finally, interest point detectors are evaluated on four datasets, and repeatability score and time of detection are used as measures of their performance.

show abstract

“…To the resulting dataset we also added images distorted with Gaussian blur (σ = 2, σ = 4). Similar transformations can be found in [10]. Finally, 200,000 images have been obtained.…”

Section: Datasetsmentioning

confidence: 75%

“…Many comparative tests have been conducted in order to evaluate state-of-the-art keypoint detection schemes [1][2][3][4][5][6][7][8][9][10]. Among measures of detectors' performance one can find time of detection, repeatability, the number of correspondences, matching score, or the number correct matches.…”

Section: Related Workmentioning

confidence: 99%

Evaluation of Interest Point Detectors in Presence of Noise

Ziomek¹,

Oszust

2016

IJISA

View full text Add to dashboard Cite

show abstract

“…For man-made object detection, it is difficult to segment the object from the background [17]. Moreover, some multi-object tracking applications, such as Digilog book, deal with many objects of different classes [21]. Humans can distinguish a mass of objects in images with petite effort, despite the fact of above described crucial situations.…”

Section: Introductionmentioning

confidence: 99%

Colour, Texture, and Shape Features based Object Recognition Using Distance Measures

Islam¹,

Pinki²

2021

IJEM

View full text Add to dashboard Cite

Object recognition is the recognizing process of objects into semantically expressive classes using its visual insides. Classification of objects becomes complex and challenging task because of its size, poor image quality, occlusion, scaling, geometric distortion, lightening, etc. In this paper, global descriptors that means Color, Texture, and Shape features are used to recognize object. Color histogram is used to obtain the color content, texture content is obtained using Gabor wavelet, and shape content is extracted using Hough transform. These low level or global features are used for creating feature vector. Distance measure is used to find the 1-Nearest Neighbor from the training images i.e. object with minimum distance or maximum similarity with visual contents of the query image. The class of that training image is the predicted label of the query image. We have used twelve different distance measures: some are metrics, some are non-metrics and finally, their recognition accuracy is compared. Ensemble of these distance measures is also used for object recognition in the image. We evaluate this method on a publicly available object-recognition dataset: Columbia Object Image Library (COIL-100) dataset. The experiments show that the recognized results outperform many state-of-the-art methods.

show abstract

“…Markers can be defined in any feature. Markers can either be 2D images [4], or feature object points [5], or take advantage of local features [6].…”

Section: Introductionmentioning

confidence: 99%

An Interaction Between User and an Augmented Reality System using A Generalized Finite State Automata and A Universal Turing Machine

Alwi*¹,

Azhari²,

Suprapto³

2019

IJITEE

View full text Add to dashboard Cite

Most interactions between users and augmented reality system (ARS) are that user assigns a marker to ARS, and the ARS responds the marker. In this context, a marker is mapped to an ARS's response, or in general, an array of markers is mapped to an array of ARS's responses. This interaction is a constant or linear complexity interaction since there is only a bijective mapping between a set of markers and a set of ARS's responses. In this research, we propose the expansion of user - ARS complexity into the polynomial. It is an interaction in which not only one marker for a single response (or an array of markers for an array of ARS's responses), but the interaction by which user provides a string of markers as a word of markers (i.e., a combination of multiple markers as a word) for a single ARS's response. The set of strings of markers to the ARS provided by users built a regular language. So that, the complexity of the user-ARS interaction became polynomial. This interaction was implemented by stating the user's language by means of a generalization of finite state automata (gFSA) and placing a universal Turing machine (UTM) between user and ARS, where the UTM as an interpreter translating or mapping the user language to ARS. To summarize our research, overall we apply the idea of a formal language into the interaction between the user and ARS, thereby changing the complexity of the interaction to polynomial even expandable to nondeterministic polynomials.

show abstract

Local Detectors and Descriptors for Object Class Recognition

Cited by 6 publications

References 25 publications

Evaluation of Interest Point Detectors in Presence of Noise

Evaluation of Interest Point Detectors in Presence of Noise

Colour, Texture, and Shape Features based Object Recognition Using Distance Measures

An Interaction Between User and an Augmented Reality System using A Generalized Finite State Automata and A Universal Turing Machine

Contact Info

Product

Resources

About