Bag-of-Words Representation in Image Annotation: A Review

Tsai, Chih‐Fong

doi:10.5402/2012/376804

Cited by 149 publications

(87 citation statements)

References 108 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In the BoW approach the SURF descriptor is additionally used to describe the salient points. This is because SURF is a well-proven point descriptor (local descriptor), widely used in BoW-based image classification processes [48]. Furthermore, SURF descriptors are based on wavelet responses, which also describe the image region in terms of textures, similar to HOG and Gabor feature descriptors.…”

Section: (B) Feature Extractionmentioning

confidence: 99%

See 1 more Smart Citation

Identification of Structurally Damaged Areas in Airborne Oblique Images Using a Visual-Bag-of-Words Approach

et al. 2016

View full text Add to dashboard Cite

Automatic post-disaster mapping of building damage using remote sensing images is an important and time-critical element of disaster management. The characteristics of remote sensing images available immediately after the disaster are not certain, since they may vary in terms of capturing platform, sensor-view, image scale, and scene complexity. Therefore, a generalized method for damage detection that is impervious to the mentioned image characteristics is desirable. This study aims to develop a method to perform grid-level damage classification of remote sensing images by detecting the damage corresponding to debris, rubble piles, and heavy spalling within a defined grid, regardless of the aforementioned image characteristics. The Visual-Bag-of-Words (BoW) is one of the most widely used and proven frameworks for image classification in the field of computer vision. The framework adopts a kind of feature representation strategy that has been shown to be more efficient for image classification-regardless of the scale and clutter-than conventional global feature representations. In this study supervised models using various radiometric descriptors (histogram of gradient orientations (HoG) and Gabor wavelets) and classifiers (SVM, Random Forests, and Adaboost) were developed for damage classification based on both BoW and conventional global feature representations, and tested with four datasets. Those vary according to the aforementioned image characteristics. The BoW framework outperformed conventional global feature representation approaches in all scenarios (i.e., for all combinations of feature descriptors, classifiers, and datasets), and produced an average accuracy of approximately 90%. Particularly encouraging was an accuracy improvement by 14% (from 77% to 91%) produced by BoW over global representation for the most complex dataset, which was used to test the generalization capability.

show abstract

Section: (B) Feature Extractionmentioning

confidence: 99%

“…Numerous feature encoding methods have been reported for visual word dictionary construction [49]. We adopted the most commonly used iterative k-means clustering algorithm [48]. The obtained feature vector is clustered into k clusters using the iterative k-means clustering [50].…”

Section: (C) Visual Words Dictionary Constructionmentioning

confidence: 99%

Identification of Structurally Damaged Areas in Airborne Oblique Images Using a Visual-Bag-of-Words Approach

et al. 2016

View full text Add to dashboard Cite

show abstract

“…For image analysis, a visual analogue of a word is used in the BoW model, which is based on the vector quantization process by clustering low-level visual features of local regions or points, such as color, texture, and so forth [18]. Representation of it uses image patches as visual words.…”

Section: Bag Of Featuresmentioning

confidence: 99%

“…SIFT and SURF can be used in image categorization. Generally, [10][11][12][13][14][15][16][17][18][19][20][21][22][23][24] better performance and efficiency of training and classification depend on better representation and clustering of features.…”

Section: Introductionmentioning

confidence: 99%

Efficient Modern Description Methods by Using SURF Algorithm for Recognition of Plant Species

Kazerouni

Schlemperz²,

Kuhnert³

2015

AIVP

View full text Add to dashboard Cite

Plants are one of one of most valuable natural resources. There would be no life on earth without plants. They unlike humans and animals manufacture their own food by photosynthesis. In every food chain, the plants occupy the first position and lead the chain as source of food. Environment and climate are largely interlinked with plants. Rainfall, humidity and temperature are influenced by presence of plants. Cutting down plants also imbalance the environment which will indirectly affect human life. Even the economic importance of plants is also quite large to mankind. Plants are great contributors of economy. Many countries rely on agriculture as one of the main source of revenue. Other benefits of plants are significant applications in different fields. Medical and agricultural applications are just some instances of plants application.Plant recognition can be done by using unique characteristic parts of plants. The used part is leaf. Shapes of leaves are useful to do plant recognition and find the species. Bag of words (BoW) and support vector machine (SVM) methods are applied to recognize and identify plants species. Visual contents of images are used and four steps are performed: (i) image preprocessing, (ii) BoW, (iii) train, (v) test. Three combined methods are used on Flavia dataset. The proposed approach is done by Speed-up robust features (SURF) method and two combined method, HARRIS-SURF and features from accelerated segment test-SURF (FAST-SURF). The accuracy of SURF method is higher than other applied methods. It is 92.28395 %. In addition to visional comparison, some quantitative results are measured and compared.

show abstract

“…In video activity recognition literature spatial information is often captured by various local space-time features as defined in [2], [3], [4], [5], [6], [7], [8], [9], [10], [11] and [12]. These local space-time features capture frame-wise spatial information by first detecting interest points with either interest point detectors (Harris detector, Hessian detectors, edge detector, corner detectors) or various sampling methods (dense sampling [13] or motion adaptive sampling [14]) for each frame, then spatio-temporal regions are defined around all the detected points in each frame and finally the spatio-temporal regions are described using one of the local space-time features.…”

Section: Introductionmentioning

confidence: 99%

Spatial information in classification of activity videos

Sengupta

Wang

Blackburn

et al. 2015

Annals of Computer Science and Information Systems

View full text Add to dashboard Cite

Abstract-Spatial information describes the relative spatial position of an object in a video. Such information may aid several video analysis tasks such as object, scene, event and activity recognition. This paper studies the effect of spatial information on video activity recognition. The paper firstly performs activity recognition on KTH and Weizmann videos using Hidden Markov Model and k-Nearest

show abstract

Bag-of-Words Representation in Image Annotation: A Review

Cited by 149 publications

References 108 publications

Identification of Structurally Damaged Areas in Airborne Oblique Images Using a Visual-Bag-of-Words Approach

Identification of Structurally Damaged Areas in Airborne Oblique Images Using a Visual-Bag-of-Words Approach

Efficient Modern Description Methods by Using SURF Algorithm for Recognition of Plant Species

Spatial information in classification of activity videos

Contact Info

Product

Resources

About