A knowledge-based approach for retrieving images by content

A knowledge based approach is introduced for retrieving images by content. It supports the answering of conceptual image queries involving similar-to predicates, spatial semantic operators, and references to conceptual terms. Interested objects in the images are represented by contours segmented from images. Image content such as shapes and spatial relationships are derived from object contours according to domain specific image knowledge. A three layered model is proposed for integrating image representations, extracted image features, and image semantics. With such a model, images can be retrieved based on the features and content specified in the queries. The knowledge based query processing is based on a query relaxation technique. The image features are classified by an automatic clustering algorithm and represented by Type Abstraction Hierarchies (TAHs) for knowledge based query processing. Since the features selected for TAH generation are based on context and user profile, and the TAHs can be generated automatically by a clustering algorithm from the feature database, our proposed image retrieval approach is scalable and context sensitive. The performance of the proposed knowledge based query processing is also discussed     

Bibliometric analysis on tendency and topics of artificial intelligence over last decade

Microsystem Technologies - Artificial intelligence (AI), together with its applications, has received world-wide attentions and is expected to exert force on the development of global economy and...     

Mixed graph convolution and residual transformation network for skeleton-based action recognition

Action recognition based on a human skeleton is an extremely challenging research problem. The temporal information contained in the human skeleton is more difficult to extract than the spatial information. Many researchers focus on graph convolution networks and apply them to action recognition. In this study, an action recognition method based on a two-stream network called RNXt-GCN is proposed on the basis of the Spatial-Temporal Graph Convolutional Network (ST-GCN). The human skeleton is converted first into a spatial-temporal graph and a SkeleMotion image which are input into ST-GCN and ResNeXt, respectively, for performing the spatial-temporal convolution. The convolved features are then fused. The proposed method models the temporal information in action from the amplitude and direction of the action and addresses the shortcomings of isolated temporal information in the ST-GCN. The experiments are comprehensively performed on the four datasets: 1) UTD-MHAD, 2) Northwestern-UCLA, 3) NTU RGB-D 60, and 4) NTU RGB-D 120. The proposed model shows very competitive results compared with other models in our experiments. On the experiments of NTU RGB?+?D 120 dataset, our proposed model outperforms those of the state-of-the-art two-stream models.

     

Semi-supervised multi-class Adaboost by exploiting unlabeled data

Semi-supervised learning has attracted much attention in pattern recognition and machine learning. Most semi-supervised learning algorithms are proposed for binary classification, and then extended to multi-class cases by using approaches such as one-against-the-rest. In this work, we propose a semi-supervised learning method by using the multi-class boosting, which can directly classify the multi-class data and achieve high classification accuracy by exploiting the unlabeled data. There are two distinct features in our proposed semi-supervised learning approach: (1) handling multi-class cases directly without reducing them to multiple two-class problems, and (2) the classification accuracy of each base classifier requiring only at least 1/K or better than 1/K (K is the number of classes). Experimental results show that the proposed method is effective based on the testing of 21 UCI benchmark data sets.     

Multiple costs based decision making with back-propagation neural networks

The current research investigates a single cost for cost-sensitive neural networks (CNN) for decision making. This may not be feasible for real cost-sensitive decisions which involve multiple costs. We propose to modify the existing model, the traditional back-propagation neural networks (TNN), by extending the back-propagation error equation for multiple cost decisions. In this multiple-cost extension, all costs are normalized to be in the same interval (i.e. between 0 and 1) as the error estimation generated in the TNN. A comparative analysis of accuracy dependent on three outcomes for constant costs was performed: (1) TNN and CNN with one constant cost (CNN-1C), (2) TNN and CNN with two constant costs (CNN-2C), and (3) CNN-1C and CNN-2C. A similar analysis for accuracy was also made for non-constant costs; (1) TNN and CNN with one non-constant cost (CNN-1NC), (2) TNN and CNN with two non-constant costs (CNN-2NC), and (3) CNN-1NC and CNN-2NC. Furthermore, we compared the misclassification cost for CNNs for both constant and non-constant costs (CNN-1C vs. CNN-2C and CNN-1NC vs. CNN-2NC). Our findings demonstrate that there is a competitive behavior between the accuracy and misclassification cost in the proposed CNN model. To obtain a higher accuracy and lower misclassification cost, our results suggest merging all constant cost matrices into one constant cost matrix for decision making. For multiple non-constant cost matrices, our results suggest maintaining separate matrices to enhance the accuracy and reduce the misclassification cost.     

Design, optimization, and performance analysis of new photodiode structures for CMOS active-pixel-sensor (APS) imager applications

The dark current in the active-pixel-sensor (APS) cell of a CMOS imager is known to be mainly generated in the regions of bird's beak after the local oxidation of silicon process as well as the surface damage caused by the implantation of high doping concentration. Furthermore, shallow and deep pn-junctions can improve the photo-sensitivity for light of short and long wavelengths, respectively. In this paper, two new photodiode structures using p-substrate and lightly-doped sensor implant SN- as pn-junction photodiode with the regions of bird's beak embraced by SN- and p-field implants, respectively, are proposed and analyzed to reduce dark current and enhance the overall spectral response. 5 /spl mu/m/spl times/5 /spl mu/m APS cells fabricated in a 0.35-/spl mu/m single-poly-triple-metal (1P3M) 3.3-V CMOS process are designed by using the proposed photodiode structures. As shown from the experimental results, the two proposed photodiode structures of 5 /spl mu/m/spl times/5 /spl mu/m APS cells have lower dark currents of 30.6 mV/s and 35.2 mV/s at the reverse-biased voltage of 2 V and higher spectral response, as compared to the conventional structure and other photodiode structures. Thus, the two proposed new photodiode structures can be applied to CMOS imager systems with small pixel size, high resolution, and high quality.     

Cathodoluminescence and Field-Emission Properties of β-Ga2O3 Nanobelts

β-Ga₂O₃ nanobelts were synthesized using a vapor transport process in a controlled ambient. Structural characterization revealed that the as-synthesized samples consisted of monoclinic β-Ga₂O₃ nanobelts, and the presence of gallium-associated defects was verified using cathodoluminescence (CL). The formation of gallium-associated defects was explained by the insufficiency of the supply of cations, generating gallium vacancies on the (010) facet during growth. Furthermore, field-emission measurements indicated that β-Ga₂O₃ nanobelts exhibited defect-related electron emission. The turn-on fields of β-Ga₂O₃ nanobelts increased significantly with the degree of structural defects. For a sample prepared under 15% ambient oxygen, Fowler–Nordheim (F–N) analysis revealed two distinct field-enhancement factors of 1194 and 276, respectively. A correlation between field emission and structural defects was proposed. The experimental results demonstrate the presence of gallium-associated defects, which behave as electron traps, degrading the electron field-emission properties of β-Ga₂O₃ nanobelts.     

Phase-shifting algorithms for electronic speckle pattern interferometry

A set of innovative phase-shifting algorithms developed to facilitate metrology based on electronic speckle pattern interferometry (ESPI) are presented. The theory of a phase-shifting algorithm, called a (5,1) algorithm, that takes five phase-shifted intensity maps before a specimen is deformed and one intensity map after a specimen is deformed is presented first. Because a high-speed camera can be used to record the dynamic image of the specimen, this newly developed algorithm has the potential to retain the phase-shifting capability for ESPI in dynamic measurements. Also shown is an algorithm called a (1,5) algorithm that takes five phase-shifted intensity maps after the specimen is deformed. In addition, a direct-correlation algorithm was integrated with these newly developed (5,1) or (1,5) algorithms to form DC-(5,1) and DC-(1,5) algorithms, which are shown to improve significantly the quality of the phase maps. The theoretical and experimental aspects of these two newly developed techniques, which can extend ESPI to areas such as high-speed dynamic measurements, are examined in detail.     

Radiation loss of grating-assisted directional couplers using the Floquet-Bloch theory

The Floquet-Bloch theory (FBT) is used to determine the radiation loss of a grating-assisted directional coupler. Improper waves obtained by FBT represent leakage and produce radiation loss along the transverse direction. By summing the fields of all leaky harmonics, the power flow radiated into the superstrate and substrate can be calculated from Poynting's theorem. However, the locations at which these summations occur impact the accuracy. An analytic formula for radiation loss is also derived. Results from both the FBT and the analytic formula are compared with results obtained using the coupled-mode theory.     

A CMOS transistor-only 8-b 4.5-Ms/s pipelined analog-to-digitalconverter using fully-differential current-mode circuit techniques

Fully-differential current-mode circuit techniques are developed for the design of a pipelined current-mode analog-to-digital converter (IADC) in the standard CMOS digital processes. In the proposed IADC, the 1-b-per-stage architecture based on the reference nonrestoring algorithm is adopted. Thus large component ratios can be avoided and the linearity errors caused by device mismatches can be minimized. As one of the key subcircuits in the IADC, an offset-canceled high speed differential current comparator (CCMP) is proposed and analyzed. In the CCMP, the subtractions of offsets are performed in the current domain without floating capacitors. Moreover, the other key subcircuit, the current sample-and-hold amplifier (CSHA), is also developed to realize the pipeline architecture. An experimental chip for the proposed IADC has been fabricated in 0.8-μm n-well CMOS technology. Using a single 5-V power supply, the fabricated IADC can be operated at 4.5-Ms/s conversion rate with a signal-to-noise-and-distortion-ratio (SNDR) of 51 db (effective 8.2-b) for the input signal at 453 kHz. For 8-b resolution, the fabricated IADC can be operated at 4.5-Ms/s conversion rate with both differential nonlinearity (DNL) and integral nonlinearity (INL) below +/-0.6 LSB. The power consumption and the active chip area are 16 mW/b and 0.73 mm²/b, respectively