Color and texture applied to a signature-based bag of visual words method for image retrieval

Santos, Joyce Miranda dos; Moura, Edleno Silva de; Silva, Altigran Soares da; Torres, Ricardo Silva

doi:10.1007/s11042-016-3955-4

Cited by 20 publications

(18 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To overcame this distance, many proposal tries to describe the visual properties of the images through numerical features (DESELAERS;KEYSERS;NEY, 2008). These features can describe visual properties of the image as a whole (TORRES;FALCAO, 2006) or they can describe each region of the image separately ZISSERMAN, 2003;SANTOS et al, 2017). However, sometimes only a part of the image contain relevant information regarding the semantic of the problem (PEREYRA et al, 2014; BLANCO et al, 2016).…”

Section: Motivationmentioning

confidence: 99%

“…Following, the Sorted Dominant Local Color and Texture (SDLCT) (SANTOS et al, 2017) is an extension of the SDLC by including textural properties of the images. In this method, not only color signatures are generated for each block, but also the signatures for their respective textures.…”

Section: Signature-based Bag-of-visual-wordsmentioning

confidence: 99%

“…Since BoVW is a visual word histogram representation of the image, it is possible to compare two images using similarity measures, such as the Cosine Angle and Jaccard distances presented on the previous Section. However, it is also possible to use textual based retrieval models such as the Vector Space Model (VSM) (SIVIC; ZISSERMAN, 2003;SANTOS et al, 2017). On the VSM, given a visual dictionary S, each image I is represented by a vector of weights W I = {w 1 , w 2 , .…”

Section: Retrieval Models Using Bovwmentioning

confidence: 99%

“…Advances in technology, such as gadgets and cell phones, enabled not only a massive capture and storage of images and videos, but also the sharing of these complex data through social media (SANTOS et al, 2017). Due to the increasing omnipresence of such data, performing decision-making in a timely manner and image retrieval tasks has been a challenge (BEDO et al, 2015a).…”

Section: Introductionmentioning

confidence: 99%

“…This can be done by using a clustering process to generate the visual words (SIVIC; ZISSERMAN, 2003) -C-BoVW. However, it is also possible to map the local features using visual signatures (SANTOS et al, 2015;SANTOS et al, 2017). In comparison to C-BoVW, the latter approach -S-BoVW -enables the identification of visual signatures at a low cost.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Managing feature extraction, mining and retrieval of complex data: applications in emergency situations and medicine

Chino¹

View full text Add to dashboard Cite

The size and complexity of the data generated by social media and medical images has increased in a fast pace. Unlike traditional data, images cannot be dealt with in its original domain, leading to rising challenges in knowledge discovery tasks. The image analysis can aid on several decision making tasks. Crowdsourcing images such as social media images can be used to increase the speed of authorities to take action in emergency situations. Images taken from the medical domain can support on daily activities of physicians to diagnose their patients. Content-Based Image Retrieval (CBIR) systems are built to retrieve similar images, being an important step for the knowledge discovery. However, in some image domains, only parts of the image are relevant to the problem. This PhD research is based on the following hypothesis: the integration of image segmentation methods with local feature CBIR system improves the precision of the retrieved images. We evaluate our proposals in two images domain: fire detection on urban emergency situations and chronic skin ulcer images. The main contributions of this PhD research can be divided in four parts. First, we propose BoWFire to detect and segment fire in emergency situations. We explore the combination of color and texture features through superpixels to detect fire in still images. Then, we explore the use of superpixels to extract local features with BoSS. BoSS is a Bag-of-Visual-Words (BoVW) approach based on visual signatures. To integrate segmentation methods with CBIR, we propose ICARUS, a skin ulcer image retrieval framework. ICARUS integrate segmentations methods based on superpixels with BoVW. We also propose ASURA, a deep learning segmentation method for skin ulcer lesions. Besides segmenting skin ulcer lesions, ASURA is able to estimate the area of the lesion in real-world units by analyzing real-world objects present in the images. Our experiments show that our proposals achieved a better precision while retrieving the most similar images in comparison with the existing approaches.

show abstract