Logo detection with extendibility and discrimination

Logos are specially designed marks that identify goods, services, and organizations using distinguished characters, graphs, signals, and colors. Identifying logos can facilitate scene understanding, intelligent navigation, and object recognition. Although numerous logo recognition methods have been proposed for printed logos, a few methods have been specifically designed for logos in photos. Furthermore, most recognition methods use codebook-based approaches for the logos in photos. A codebook-based method is concerned with the generation of visual words for all the logo models. When new logos are added, the codebook reconstruction is required if effectiveness is a crucial factor. Moreover, logo detection in natural scenes is difficult because of perspective tilt and non-rigid deformation. Therefore, this study develops an extendable, but discriminating, model-based logo detection method. The proposed logo detection method is based on a support vector machine (SVM) using edge-based histograms of oriented gradient (HOGE) as features through multi-scale sliding window scanning. Thereafter, anti-distortion affine scale invariant feature transform (ASIFT) is used for logo verification with constraints on the ASIFT matching pairs and neighbors. The experimental results using the public Flickr-Logo database confirm that the proposed method has a higher retrieval and precision accuracy compared to existing model-based methods.

     

Optimal entropy-constrained non-uniform scalar quantizer design for low bit-rate pixel domain DVC

In this paper, an optimal entropy-constrained non-uniform scalar quantizer is proposed for the pixel domain DVC. The uniform quantizer is efficient for the hybrid video coding since the residual signals conforming to a single-variance Laplacian distribution. However, the uniform quantizer is not optimal for pixel domain distributed video coding (DVC). This is because the uniform quantizer is not adaptive to the joint distribution of the source and the SI, especially for low level quantization. The signal distribution of pixel domain DVC conforms to the mixture model with multi-variance. The optimal non-uniform quantizer is designed according to the joint distribution, the error between the source and the SI can be decreased. As a result, the bit rate can be saved and the video quality won’t sacrifice too much. Accordingly, a better R-D trade-off can be achieved. First, the quantization level is fixed and the optimal RD trade-off is achieved by using a Lagrangian function J(Q). The rate and distortion components is designed based on P(Y|Q). The conditional probability density function of SI Y depend on quantization partitions Q, P(Y|Q), is approximated by a Guassian mixture model at encocder. Since the SI can not be accessed at encoder, an estimation of P(Y|Q) based on the distribution of the source is proposed. Next, J(Q) is optimized by an iterative Lloyd-Max algorithm with a novel quantization partition updating algorithm. To guarantee the convergence of J(Q), the monotonicity of the interval in which the endpoints of the quantizer lie must be satisfied. Then, a quantizer partition updating algorithm which considers the extreme points of the histogram of the source is proposed. Consequently, the entropy-constrained optimal non-uniform quantization partitions are derived and a better RD trade-off is achieved by applying them. Experiment results show that the proposed scheme can improve the performance by 0.5 dB averagely compared to the uniform scalar quantization.     

Coverage enhancement by using the mobility of mobile sensor nodes

Coverage is a fundamental problem in sensor networks. Sensor coverage, which reflects how well a sensor network is monitored by sensors, is an important measure for the quality of service (QoS) that a sensor network can provide. In mobile sensor networks, the mobility of sensor nodes can be utilized to enhance the coverage of the network. Since the movement of sensor nodes will consume much energy, this mobility of sensor nodes should be properly managed by some pre-defined schemes or protocols. By noticing this issue, some existing works have proposed several movement-assisted sensor deployment schemes. These works assume that the target field is a 2-dimensional space. In this paper, we study a generalized case of this problem whereby the target field can be a space which ranges from 1-dimensional to 3-dimensional. Two variations of the movement-assisted sensor deployment problem with different optimization objectives were formulated. We identify a set of basic attributes which can be used as guidelines for designing movement-assisted sensor deployment schemes. Based on these attributes, we propose efficient algorithms for both variants of the movement-assisted sensor deployment problem.     

Moving people tracking with detection by latent semantic analysis for visual surveillance applications

The latent semantic analysis (LSA) has been widely used in the fields of computer vision and pattern recognition. Most of the existing works based on LSA focus on behavior recognition and motion classification. In the applications of visual surveillance, accurate tracking of the moving people in surveillance scenes, is regarded as one of the preliminary requirement for other tasks such as object recognition or segmentation. However, accurate tracking is extremely hard under challenging surveillance scenes where similarity among multiple objects or occlusion among multiple objects occurs. Usual temporal Markov chain based tracking algorithms suffer from the ‘tracking error accumulation problem’. The accumulated errors can finally make the tracking to drift from the target. To handle the problem of tracking drift, some authors have proposed the idea of using detection along with tracking as an effective solution. However, many of the critical issues still remain unsettled in these detection based tracking algorithms. In this paper, we propose a novel moving people tracking with detection based on (probabilistic) LSA. By employing a novel ‘twin-pipeline’ training framework to find the latent semantic topics of ‘moving people’, the proposed detection can effectively detect the interest points on moving people in different indoor and outdoor environments with camera motion. Since the detected interest points on different body parts can be used to locate the position of moving people more accurately, by combining the detection with incremental subspace learning based tracking, the proposed algorithms resolves the problem of tracking drift during each target appearance update process. In addition, due to the time independent processing mechanism of detection, the proposed method is also able to handle the error accumulation problem. The detection can calibrate the tracking errors during updating of each state of the tracking algorithm. Extensive, experiments on various surveillance environments using different benchmark datasets have proved the accuracy and robustness of the proposed tracking algorithm. Further, the experimental comparison results clearly show that the proposed tracking algorithm outperforms the well known tracking algorithms such as ISL, AMS and WSL algorithms. Furthermore, the speed performance of the proposed method is also satisfactory for realistic surveillance applications.     

浅谈玻璃钢的加工

本文在实际研究和应用的基础上，对玻璃钢的传统加工方法和特种加工方法，如激光切割、纯水射流切割和磨料水射流切割作了简要论述。     

Online mining abnormal period patterns from multiple medical sensor data streams

With the advanced technology of medical devices and sensors, an abundance of medical data streams are available. However, data analysis techniques are very limited, especially for processing massive multiple physiological streams that may only be understood by medical experts. The state-of-the-art techniques only allow multiple medical devices to independently monitor different physiological parameters for the patient’s status, thus they signal too many false alarms, creating unnecessary noise, especially in the Intensive Care Unit (ICU). An effective solution which has been recently studied is to integrate information from multiple physiologic parameters to reduce alarms. But it is a challenge to detect abnormalities from high frequently changed physiological streams data, since abnormalities occur gradually due to the complex situation of patients. An analysis of ICU physiological data streams shows that many vital physiological parameters are changed periodically (such as heart rate, arterial pressure, and respiratory impedance) and thus abnormalities are generally abnormal period patterns. In this paper, we develop a Mining Abnormal Period Patterns from Multiple Physiological Streams (MAPPMPS) method to detect and rank abnormalities in medical sensor streams. The efficiency and effectiveness of the MAPPMPS method is demonstrated by a real-world massive database of multiple physiological streams sampled in ICU, comprising 250 patients’ streams (each stream involving over 1.3 million data points) with a total size of 28 GB data.     

Discovering small-world in association link networks for association learning

Association Link Network (ALN) is a kind of Semantic Link Network built by mining the association relations among multimedia Web resources for effectively supporting Web intelligent application such as Web-based learning, and semantic search. This paper explores the Small-World properties of ALN to provide theoretical support for association learning (i.e., a simple idea of “learning from Web resources”). First, a filtering algorithm of ALN is proposed to generate the filtered status of ALN, aiming to observe the Small-World properties of ALN at given network size and filtering parameter. Comparison of the Small-World properties between ALN and random graph shows that ALN reveals prominent Small-World characteristic. Then, we investigate the evolution of Small-World properties over time at several incremental network sizes. The average path length of ALN scales with the network size, while clustering coefficient of ALN is independent of the network size. And we find that ALN has smaller average path length and higher clustering coefficient than WWW at the same network size and network average degree. After that, based on the Small-World characteristic of ALN, we present an Association Learning Model (ALM), which can efficiently provide association learning of Web resources in breadth or depth for learners.     

The min-dist location selection and facility replacement queries

We propose and study a new type of location optimization problem, the min-dist location selection problem: given a set of clients and a set of existing facilities, we select a location from a given set of potential locations for establishing a new facility, so that the average distance between a client and her nearest facility is minimized. The problem has a wide range of applications in urban development simulation, massively multiplayer online games, and decision support systems. We also investigate a variant of the problem, where we consider replacing (instead of adding) a facility while achieving the same optimization goal. We call this variant the min-dist facility replacement problem. We explore two common approaches to location optimization problems and present methods based on those approaches for solving the min-dist location selection problem. However, those methods either need to maintain an extra index or fall short in efficiency. To address their drawbacks, we propose a novel method (named MND), which has very close performance to the fastest method but does not need an extra index. We then utilize the key idea behind MND to approach the min-dist facility replacement problem, which results in two algorithms names MSND and RID. We provide a detailed comparative cost analysis and conduct extensive experiments on the various algorithms. The results show that MND and RID outperform their competitors by orders of magnitude.     

An FAR-SW based approach for webpage information extraction

Automatically identifying and extracting the target information of a webpage, especially main text, is a critical task in many web content analysis applications, such as information retrieval and automated screen reading. However, compared with typical plain texts, the structures of information on the web are extremely complex and have no single fixed template or layout. On the other hand, the amount of presentation elements on web pages, such as dynamic navigational menus, flashing logos, and a multitude of ad blocks, has increased rapidly in the past decade. In this paper, we have proposed a statistics-based approach that integrates the concept of fuzzy association rules (FAR) with that of sliding window (SW) to efficiently extract the main text content from web pages. Our approach involves two separate stages. In Stage 1, the original HTML source is pre-processed and features are extracted for every line of text; then, a supervised learning is performed to detect fuzzy association rules in training web pages. In Stage 2, necessary HTML source preprocessing and text line feature extraction are conducted the same way as that of Stage 1, after which each text line is tested whether it belongs to the main text by extracted fuzzy association rules. Next, a sliding window is applied to segment the web page into several potential topical blocks. Finally, a simple selection algorithm is utilized to select those important blocks that are then united as the detected topical region (main texts). Experimental results on real world data show that the efficiency and accuracy of our approach are better than existing Document Object Model (DOM)-based and Vision-based approaches.     

Social media assimilation in firms: Investigating the roles of absorptive capacity and institutional pressures

Firms are increasingly employing social media to manage relationships with partner organizations, yet the role of institutional pressures in social media assimilation has not been studied. We investigate social media assimilation in firms using a model that combines the two theoretical streams of IT adoption: organizational innovation and institutional theory. The study uses a composite view of absorptive capacity that includes both previous experience with similar technology and the general ability to learn and exploit new technologies. We find that institutional pressures are an important antecedent to absorptive capacity, an important measure of organizational learning capability. The paper augments theory in finding the role and limits of institutional pressures. Institutional pressures are found to have no direct effect on social media assimilation but to impact absorptive capacity, which mediates its influence on assimilation.