Content Classification of Multimedia Documents using Partitions of Low-Level Features

Audio-visual documents obtained from German TV news are classified according to the IPTC topic categorization scheme. To this end usual text classification techniques are adapted to speech, video, and non-speech audio. For each of the three modalities word analogues are generated: sequences of syllables for speech, “video words” based on low level color features (color moments, color correlogram and color wavelet), and “audio words” based on low-level spectral features (spectral envelope and spectral flatness) for non-speech audio. Such audio and video words provide a means to represent the different modalities in a uniform way. The frequencies of the word analogues represent audio-visual documents: the standard bag-of-words approach. Support vector machines are used for supervised classification in a 1 vs. n setting. Classification based on speech outperforms all other single modalities. Combining speech with non-speech audio improves classification. Classification is further improved by supplementing speech and non-speech audio with video words. Optimal F-scores range between 62% and 94% corresponding to 50% - 84% above chance. The optimal combination of modalities depends on the category to be recognized. The construction of audio and video words from low-level features provide a good basis for the integration of speech, non-speech audio and video.

Analysis of the Distributions of Color Characteristics in Art Painting Images

In this paper we study some of the characteristics of the art painting image color semantics. We analyze the color features of differ- ent artists and art movements. The analysis includes exploration of hue, saturation and luminance. We also use quartile’s analysis to obtain the dis- tribution of the dispersion of defined groups of paintings and measure the degree of purity for these groups. A special software system “Art Paint- ing Image Color Semantics” (APICSS) for image analysis and retrieval was created. The obtained result can be used for automatic classification of art paintings in image retrieval systems, where the indexing is based on color characteristics.

Hand detection on images based on deformable part models and additional features

Hand detection on images has important applications on person activities recognition. This thesis focuses on PASCAL Visual Object Classes (VOC) system for hand detection. VOC has become a popular system for object detection, based on twenty common objects, and has been released with a successful deformable parts model in VOC2007. A hand detection on an image is made when the system gets a bounding box which overlaps with at least 50% of any ground truth bounding box for a hand on the image. The initial average precision of this detector is around 0.215 compared with a state-of-art of 0.104; however, color and frequency features for detected bounding boxes contain important information for re-scoring, and the average precision can be improved to 0.218 with these features. Results show that these features help on getting higher precision for low recall, even though the average precision is similar.

A convolutional neural network for automatic analysis of aerial imagery

This paper introduces a new method to automate the detection of marine species in aerial imagery using a Machine Learning approach. Our proposed system has at its core, a convolutional neural network. We compare this trainable classifier to a handcrafted classifier based on color features, entropy and shape analysis. Experiments demonstrate that the convolutional neural network outperforms the handcrafted solution. We also introduce a negative training example-selection method for situations where the original training set consists of a collection of labeled images in which the objects of interest (positive examples) have been marked by a bounding box. We show that picking random rectangles from the background is not necessarily the best way to generate useful negative examples with respect to learning.

Memory as discrimination: what distraction reveals

Recalling information involves the process of discriminating between relevant and irrelevant information stored in memory. Not infrequently, the relevant information needs to be selected from amongst a series of related possibilities. This is likely to be particularly problematic when the irrelevant possibilities are not only temporally or contextually appropriate but also overlap semantically with the target or targets. Here, we investigate the extent to which purely perceptual features which discriminate between irrelevant and target material can be used to overcome the negative impact of contextual and semantic relatedness. Adopting a distraction paradigm, it is demonstrated that when distracters are interleaved with targets presented either visually (Experiment 1) or auditorily (Experiment 2), a within-modality semantic distraction effect occurs; semantically-related distracters impact upon recall more than unrelated distracters. In the semantically-related condition, the number of intrusions in recall is reduced whilst the number of correctly recalled targets is simultaneously increased by the presence of perceptual cues to relevance (color features in Experiment 1 or speaker’s gender in Experiment 2). However, as demonstrated in Experiment 3, even presenting semantically-related distracters in a language and a sensory modality (spoken Welsh) distinct from that of the targets (visual English) is insufficient to eliminate false recalls completely, or to restore correct recall to levels seen with unrelated distracters . Together, the study shows how semantic and non-semantic discriminability shape patterns of both erroneous and correct recall.

Método de detecção e classificação de lesões de pele em imagens digitais a partir do modelo Chan-Vese e máquina de vetor de suporte

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Crenicichla chicha, a new species of pike cichlid (Teleostei: Cichlidae) from the rio Papagaio, upper rio Tapajos basin, Mato Grosso, Brazil

Crenicichla chicha, new species, occurs in clear, fast-running waters with rocky substrates in the rio Papagaio and tributaries. It is distinguished from all other Crenicichla species by the combination of two character states: infraorbitals 3 and 4 co-ossified (vs. separated) and 66-75 scales in the row immediately above to that containing the lower lateral line (E1 row scales). Crenicichla chicha shares a smooth preopercular margin, co-ossification of infraorbitals 3 and 4, and some color features with C. hemera from the adjacent rio Aripuana drainage, rio Madeira basin. It differs from Crenicichla hemera in more E1 scales (6675 vs. 58-65) and presence of a conspicuous black narrow stripe running from infraorbital 3 obliquely caudoventrad toward the preopercular margin vs. a rounded and faint suborbital marking present on infraorbitals 3-4. Examination of the type series and additional material from the rio Aripuana confirms that Crenicichla guentheri Ploeg, 1991 is a junior subjective synonym of C. hemera Kullander, 1990.

Crenicichla chicha, a new species of pike cichlid (Teleostei: Cichlidae) from the rio Papagaio, upper rio Tapajós basin, Mato Grosso, Brazil

Crenicichla chicha, new species, occurs in clear, fast-running waters with rocky substrates in the rio Papagaio and tributaries. It is distinguished from all other Crenicichla species by the combination of two character states: infraorbitals 3 and 4 co-ossified (vs. separated) and 66-75 scales in the row immediately above to that containing the lower lateral line (E1 row scales). Crenicichla chicha shares a smooth preopercular margin, co-ossification of infraorbitals 3 and 4, and some color features with C. hemera from the adjacent rio Aripuanã drainage, rio Madeira basin. It differs from Crenicichla hemera in more E1 scales (66-75 vs. 58-65) and presence of a conspicuous black narrow stripe running from infraorbital 3 obliquely caudoventrad toward the preopercular margin vs. a rounded and faint suborbital marking present on infraorbitals 3-4. Examination of the type series and additional material from the rio Aripuanã confirms that Crenicichla guentheri Ploeg, 1991 is a junior subjective synonym of C. hemera Kullander, 1990.

Fuzzy clustering of color and texture features for image segmentation: a study on satellite image retrieval

In this paper, a new image segmentation approach that integrates color and texture features using the fuzzy c-means clustering algorithm is described. To demonstrate the applicability of the proposed approach to satellite image retrieval, an interactive region-based image query system is designed and developed. A database comprising 400 multispectral satellite images is used to evaluate the performance of the system. The results are analyzed and discussed, and a performance comparison with other methods is included. The outcomes reveal that the proposed approach is able to improve the quality of the segmentation results as well as the retrieval performance.

The capitals of the globe : the political, commercial, artistic and sacred capitals of Europe, Asia, Africa, North America, South America and the West Indies with graphic and accurate descriptions of all their wonders, shrines, traditions, sensations and treasures ; their political and commercial importance ; their religious, artistic and climatic attractions ; their geographical positions, physical features, history, population, industries, trade, traffic, currency, communications and hotels : the manners and customs of the inhabitants-types and costumes together with every information how to reach them, how to travel, and what it costs : superbly and profusely illustrated with engravings and color maps /

pt. 1. The capitals of Europe -- pt. 2. The capitals of Asia -- pt. 3. The capitals of Africa -- pt. 4. The capitals of North America -- pt. 5. The capitals of Central America, South America and the West Indes.

Color and texture feature fusion using kernel PCA with application to object-based vegetation species classification

A good object representation or object descriptor is one of the key issues in object based image analysis. To effectively fuse color and texture as a unified descriptor at object level, this paper presents a novel method for feature fusion. Color histogram and the uniform local binary patterns are extracted from arbitrary-shaped image-objects, and kernel principal component analysis (kernel PCA) is employed to find nonlinear relationships of the extracted color and texture features. The maximum likelihood approach is used to estimate the intrinsic dimensionality, which is then used as a criterion for automatic selection of optimal feature set from the fused feature. The proposed method is evaluated using SVM as the benchmark classifier and is applied to object-based vegetation species classification using high spatial resolution aerial imagery. Experimental results demonstrate that great improvement can be achieved by using proposed feature fusion method.

Satellite evidence of harmful algal blooms and related oceanographic features in the Bohai Sea during autumn 1998

Harmful algal blooms (HABs) are truly global marine phenomena of increasing significance. Some HAB occurrences are different to observe because of their high spatial and temporal variability and their advection, once formed, by surface currents. A serious HAB occurred in the Bohai Sea during autumn 1998, causing the largest fisheries economic loss. The present study analyzes the formation, distribution, and advection of HAB using satellite SeaWiFS ocean color data and other oceanographic data. The results show that the bloom originated in the western coastal waters of the Bohai Sea in early September, and developed southeastward when sea surface temperature (SST) increased to 25-26 °C. The bloom with a high Chl-a concentration (6.5 mg m-3) in center portion covered an area of 60 × 65 km2. At the end of September, the bloom decayed when SST decreased to 22-23 °C. The HAB may have been initiated by a combination of the river discharge nutrients in the west coastal waters and the increase of SST; afterwards it may have been transported eastward by the local circulation that was enhanced by northwesterly winds in late September and early October.

Visual search near threshold: Some features are more equal than others.

While searching for objects, we combine information from multiple visual modalities. Classical theories of visual search assume that features are processed independently prior to an integration stage. Based on this, one would predict that features that are equally discriminable in single feature search should remain so in conjunction search. We test this hypothesis by examining whether search accuracy in feature search predicts accuracy in conjunction search. Subjects searched for objects combining color and orientation or size; eye movements were recorded. Prior to the main experiment, we matched feature discriminability, making sure that in feature search, 70% of saccades were likely to go to the correct target stimulus. In contrast to this symmetric single feature discrimination performance, the conjunction search task showed an asymmetry in feature discrimination performance: In conjunction search, a similar percentage of saccades went to the correct color as in feature search but much less often to correct orientation or size. Therefore, accuracy in feature search is a good predictor of accuracy in conjunction search for color but not for size and orientation. We propose two explanations for the presence of such asymmetries in conjunction search: the use of conjunctively tuned channels and differential crowding effects for different features.

A versatile color tuning strategy for iridium(III) and platinum(II) electrophosphors by shifting the charge-transfer states with an electron-deficient core

By fusing an electron-deficient ring system with the phenyl ring of a 2-phenylpyridine (ppy)-type ligand, a new and synthetically versatile strategy for the phosphorescence color tuning of cyclometalated iridium(III) and platinum(II) metallophosphors has been established. Two robust red electrophosphors with enhanced electron-injection/electron-transporting features were prepared by using an electron-trapping fluoren-9-one chromophore in the ligand design. The thermal, photophysical, redox and electrophosphorescent properties of these complexes are reported. These exciting results can be attributed to a switch of the metal-to-ligand charge-transfer (MLCT) character of the transition from the pyridyl groups in the traditional Ir-III or Pt-II ppy-type complexes to the electron-deficient ring core, and the spectral assignments corroborate well with the electrochemical data as well as the timedependent density functional theory (TD-DFT) calculations. The electron-withdrawing character of the fused ring results in much more stable MLCT states, inducing a substantial red-shift of the triplet emission energy from yellow to red for the Ir-III complex and even green to red for the PtII counterpart.