Search

Purpose This paper aims to build a legal intelligent auxiliary discretionary system for predicting the penalty and damage compensation values. After extensively considering current the characteristics of the current Chinese legal system, a practical legal intelligent auxiliary discretionary system based on genetic algorithm-backpropagation (GA-BP) neural network (NN) is proposed herein. Design/methodology/approach An experiment is designed to analyze cases involving mental anguish compensation in medical disputes, and a Chinese legal intelligent auxiliary discretionary adviser system is built based on a GA-BP NN. Because BP neural networks perform well for nonlinear problems and GAs can improve their ability to find optimal values, and accelerate their convergence, a combined GA–BP algorithm is used. In addition, an ontology is used to reduce the semantic ambiguities and extract the implied semantic information. Findings We confirm that a case-based legal intelligent auxiliary discretionary adviser system based on a GA-BP NN and ontology techniques has good performance in prediction. By predicting the mental anguish compensation values, the legal intelligent auxiliary discretionary adviser system can help judges to handle cases more quickly and ordinary people to discover the suggested compensation or penalty. In contrast to BP NN or SVM, the result seems more close to the actual compensation rate. Practical implications Recently, smart court has been developed in China; the purpose of which is to build the legal advice system for improving judicial justice and reducing differences in sentencing. A practical legal advice system is an urgent requirement for the judiciary. Originality/value This paper presents a study of a case-based legal intelligent auxiliary discretionary adviser system based on a GA-BP NN and ontology techniques. The findings offer advice to optimize legal intelligent auxiliary discretionary adviser systems for mental anguish compensation in medical disputes.

This paper is a summary of the author's research process in the developing and designing the image recognition system based on artificial intelligence. ARTIFICIAL INTELLIGENCE CONCEPTS •Image recognition refers to the technology of using computer to process, analyze and understand the image to identify different patterns of targets and objects. It is a practical application of deep learning algorithm. At present, image recognition technology is generally divided into face recognition and commodity recognition. Face recognition is mainly used in security inspection, identity verification and mobile payment. Commodity recognition is mainly used in commodity circulation process, especially in unmanned retail fields such as unmanned shelf and intelligent retail cabinet.

We study the two-echelon vehicle routing problem, in which the delivery to customers from a depot with transits through satellites. The goal is to find the minimum total routing cost and minimize the cost of single point of demand, more precisely, to find an approximate algorithm minimizing the approximation ratio, which compares the performance of this approximate algorithm with the optimal solution. In this paper, we describe a formulation for this problem and present an approximate algorithm GA to find the minimum total routing cost. We prove approximation ratio for the algorithm GA. Finally, we conduct extensive computational experiments to measure the performance of the approximate algorithm GA, the computational results indicate that the approximate algorithm GA can effectively solve various instances of the problem.

Polar codes are proved to be able to theoretically achieve the Shannon limit. However, the performance of polar codes with short code length is not well in practice. One widely used method to improve the short length codes is concatenation. Recently, staircase coding structure provides an efficient concatenation scheme for finite length block codes, which the component block code can concatenate itself to improve the coding performance. Thus, in this paper, we propose a high-rate polar-staircase coding scheme with systematic polar codes as the component codes. The polar-staircase coding scheme can enhance the unreliable parts of the polar codes through the concatenation. To achieve the asymptotic performance, which is mainly depending on the decoding algorithms, three soft decoding algorithms are analyzed for our polar-staircase coding. We first investigate the conventional belief propagation (BP) decoding and soft cancellation (SCAN) decoding. The performance of the two algorithms is not well in the short length regime. Then, we adopt and optimize a soft successive cancellation list (SSCL) decoding algorithm for the polar-staircase codes with the tradeoff between reliability and complexity. Simulations show that the SSCL decoding outperforms than the other soft decoding algorithms over the AWGN channels.

Synthetic aperture radar (SAR) image despeckling is an essential problem in remote sensing technology, which has a strong influence on the performance of the following processing. We propose a new despeckled algorithm combining CNN and fractional-order total variation. Through constructing a CNN model and introducing the fractional-order total variation into loss function as the regularization term, the experimental results prove that our proposed method can avoid detail ambiguity and overly smooth caused by integral-order, and preserve rich texture and details information. Therefore, the high-quality despeckled images generated by our model will significantly improve the availability of SAR images.

The long-haul communication systems can offer ultra high-speed data transfer rates but suffer from burst errors. The high-rate and high-performance staircase codes provide an efficient way for long-haul transmission. The staircase coding scheme is a concatenation structure, which provides the opportunity to improve the performance of high-rate polar codes. At the same time, the polar codes make the staircase structure more reliable. Thus, a high-rate polar-staircase coding scheme is proposed, where the systematic polar codes are applied as the component codes. The soft cancellation decoding of the systematic polar codes is proposed as a basic ingredient. The encoding of the polar-staircase codes is designed with the help of density evolution, where the unreliable parts of the polar codes are enhanced. The corresponding decoding is proposed with low complexity, and is also optimized for burst error channels. With the well designed encoding and decoding algorithms, the polar-staircase codes perform well on both AWGN channels and burst error channels., Comment: 6 pages, 4 figures. Accepted for publication at IEEE Vehicular Technology Conference (VTC), Fall 2018

An unstructured grid adaptation technology has been developed for capturing flow features effectively. The face-based hierarchical data structure is used at grid refinement procedure. The difference of suitable flow properties between cell and its neighbors for flow features detection. The results of different simulation cases show the strengths of unstructured grid adaptation technology. The flow features such as shock wave and vortex are resolved better with grids refinement. The grid adaptation also improves the computational accuracy of aerodynamic characteristics and loads.

n order to support design–manufacturing integration (DMI), a transformation method is proposed by recognizing the features of EBOM and MBOM, to effectively ensure data integrity, correctness and consistency in systems. By defining different components, the transformation processes are identified and a formal transformation model of BOM view is proposed. Then, rules for BOM transforming and conversion algorithms are described. Engineering change problems also are solved by processing rules. This research has achieved an automatic conversion of BOM. Finally, the practical application in the enterprise indicated that this method is effective for DMI.

Taller people are more accurate in judging target location in the intermediate distance range., Our sense of vision reliably directs and guides our everyday actions, such as reaching and walking. This ability is especially fascinating because the optical images of natural scenes that project into our eyes are insufficient to adequately form a perceptual space. It has been proposed that the brain makes up for this inadequacy by using its intrinsic spatial knowledge. However, it is unclear what constitutes intrinsic spatial knowledge and how it is acquired. We investigated this question and showed evidence of an ecological basis, which uses the statistical spatial relationship between the observer and the terrestrial environment, namely, the ground surface. We found that in dark and reduced-cue environments where intrinsic knowledge has a greater contribution, perceived target location is more accurate when referenced to the ground than to the ceiling. Furthermore, taller observers more accurately localized the target. Superior performance was also observed in the full-cue environment, even when we compensated for the observers’ heights by having the taller observer sit on a chair and the shorter observers stand on a box. Although fascinating, this finding dovetails with the prediction of the ecological hypothesis for intrinsic spatial knowledge. It suggests that an individual’s accumulated lifetime experiences of being tall and his or her constant interactions with ground-based objects not only determine intrinsic spatial knowledge but also endow him or her with an advantage in spatial ability in the intermediate distance range.

Attention readily facilitates the detection and discrimination of objects, but it is not known whether it helps to form the vast volume of visual space that contains the objects and where actions are implemented. Conventional wisdom suggests not, given the effortless ease with which we perceive three-dimensional (3D) scenes on opening our eyes. Here, we show evidence to the contrary. In Experiment 1, the observer judged the location of a briefly presented target, placed either on the textured ground or ceiling surface. Judged location was more accurate for a target on the ground, provided that the ground was visible and that the observer directed attention to the lower visual field, not the upper field. This reveals that attention facilitates space perception with reference to the ground. Experiment 2 showed that judged location of a target in mid-air, with both ground and ceiling surfaces present, was more accurate when the observer directed their attention to the lower visual field; this indicates that the attention effect extends to visual space above the ground. These findings underscore the role of attention in anchoring visual orientation in space, which is arguably a primal event that enhances one's ability to interact with objects and surface layouts within the visual space. The fact that the effect of attention was contingent on the ground being visible suggests that our terrestrial visual system is best served by its ecological niche.

Personal and network encryption application models nowadays are analyzed in this paper, then a team secrecy encryption application model is proposed based on practical applications. In this model, members of one team have same access authority for electronic documents, and team is the smallest unit of key distribution. Procedures of distribution, transmission and updating key are offline. This model is suitable for co-confidential application among many departments which are loose coupling. An encryption system is implemented based on this model aimed at protecting Microsoft office document.

We design and realize a testing and analyzing tool for performance of image compression algorithms. This tool can test and analyze compression algorithms and generate kinds of analysis chart automatically, provides a lot of convenience for users and has very important practical value.

We provide a sweep coverage algorithm for routing mobile sensors that communicate with a central data sink. This algorithm improves on its predecessors by reducing the number of unnecessary scans when different points of interest (POIs) have different requirements for the time interval within which they must be scanned (sweep period). Most sweep coverage algorithms seek to minimize the number of sensors required to cover a given collection of POIs. When POIs have different sweep period requirements, existing algorithms will produce solutions in which sensors visit some POIs much more frequently than is necessary. We define this as the POI Over-Coverage problem. In order to address this problem we develop a Periodic Sweep Coverage (PSC) scheme based on a well-known solution to the Periodic Vehicle Routing Problem (PVRP). Our algorithm seeks a route for the mobile sensors that minimizes the number of unnecessary visits to each POI. To verify and test the proposed scheme we implemented a C++ simulation and ran scenarios with a variety of POI topologies (number and distribution of the POIs) and the speed at which sensors could travel. The simulation results show that the PSC algorithm outperforms other sweep coverage algorithms such as CSweep and Vehicle Routing Problem Sweep Coverage (VRPSC) on both the average number of sensors in a solution and in the computational time required to find a solution. Our results also demonstrate that the PSC scheme is more suitable for the sweep coverage scenarios in which higher speed mobile sensors are used.

With the fast development of social economy, the competition between enterprises in construction projects becomes more and more serious. The cost control method is proposed to the bidding process of enterprises based on the gray system forecast. At the same time, the cost control software is also constructed to the bidding process of enterprises. Experimental results suggest that the proposed approach is feasible and correct.

An important element of BBS public opinion analysis system is to update the contents of knowledge regular library dynamically adding sensitive words constantly to knowledge regular library. Apriori algorithm is classical algorithm of mining association rules. This paper presents a new method of adding the contents of knowledge regular automatically using association rule models. Experimental results show that the system obviously improves the accuracy of monitoring BBS public opinion by means of this algorithm.

According to technical statistics, current TCP protocols with approximately 80% Internet applications run on perform very well on wired networks. However, due to the effects of long propagation delay, great bandwidth asymmetry, high sporadic Bit Error Rate (BER) and etc., TCP performance degrades obviously on the satellite communication networks. To avoid the problems, TP-S, a novel transport control protocol, is introduced for satellite IP networks. Firstly, in order to increase the increment speed of Congestion Window (cwnd) at the beginning of data transmission, the traditional Slow Start strategy is replaced by a new strategy, known as Super Start. Secondly, a new packet lost discriminated scheme based on IP packets alternately sending with different priority is used in the protocol to decouple congestion decision from errors. Thirdly, bandwidth asymmetry problem is avoided by adopting Modified NACK (M-NACK) in receiving ends, which is sent periodically. In addition, the sending strategy in routers is also modified along with other's changes to support the protocol. Finally, the simulation experiments show that the new protocol can not only significantly enhance throughput performance, but also reduce sharply bandwidth used in the reverse path as compared with traditional TCP protocols and those protocols, which are recently proposed for satellite IP networks.

The traditional Gabor transform need calculate convolution with whole image when extract expression feature. So, the feature dimensions are very large and computation complexity is very high. Though some improved methods can reduce the feature dimensions and computing cost by manually marking out the feature points, but those methods need intense human intervention and can not meet to the need of automatic recognition. In This paper, a new face expression recognition method are proposed based on feature fusion which combines the Gabor transform and ASM automatic feature orientation technology. Firstly, ASM technology is used to locate the feature point of a face shape. And then Gabor feature of feature point are extracted by using Gabor transform. Finally, the two feature sets are fused to implement the face expression recognition. The experimental results show that the proposed method can effectively utilize the local texture information and global shape information of expression and get better recognition effect.

Aiming at the problem in facial feature location, the Active Shape Models algorithm has no evaluable criterion to judge its convergence, and tend to run into local minimum when using the fixed search scale, and how to choose the scale is always an issue. This paper presents a method based on the whole shape gray grads to evaluate the search effects, and an improved search policy of multiscale. The experiment results show that, this method can enhance the robustness of ASM feature location effectively and reduce the computational complexity.

In this paper, a novel feature extraction approach alpha-based supervised orthogonal projection reduction by affinity is proposed by introducing the idea of SLLE into the traditional method of OPRA. By adding an additional parameter a to control the degree of supervision, the proposed method can acquire some compromise between purely supervised OPRA and unsupervised OPRA and does not only keep the reservation of some flow-shaped structure during high-dimensional to low-dimensional mapping, but also gets better orthogonal projection. Experimental results based on both synthetic data and real data (human face recognition) show that the proposed method is more effective than either purely supervised OPRA or unsupervised OPRA and some other traditional feature extraction methods.

Information about the impending collision of an approaching object can be specified by visual and auditory means. We examined the discrimination thresholds for vision, audition, and vision/audition combined, in the processing of time-to-collision (TTC) of an approaching object. The stimulus consisted of a computer simulated car approaching on a flat ground towards the participants which disappeared at a certain point before collision. After the presentation of two approaching movements in succession, participants pressed a button to indicate which of the two movements would result in the car colliding with the viewpoint sooner from the moment it disappeared. The results demonstrated that most participants were sensitive to TTC information provided by a visual source, but not when provided by an auditory source. That said, auditory information provided effective static distance information. When both sources of information were combined, participants used the most accurate source of information to make their judgments.

Albrrad-Multilay erdynsm*sarsocia tivenwraln~ork~~ is a spa.cjoint random architeeturr with feedback The neees~~ly for the existem of the architffhup show that the research for synaptic weight haining ip in need. In the Gmt, the pap analyzes the differenoe between Inultilayer dynamics dtive network with double layer associative nymo'y network, BS well as the resemblance with discrete Hopfield nehvork by physics model Then, it proves the stability of !earning Plgorithm of multilayer dynamics associatiVe nehvork based on genernlhed Hebb rule in mathematics The 6mulahi.e experiment for the algorithm getr 6ne result and SOW a fundamental Bue of muhilayer .dynamiw dhe nerve in enginewing praciicea and researeh in "ly. IkTm-MukiLpw-eGenemlired Hebb rule, Discmte Hopfipldnehwrk, SynrIpbc wpigks Lyqunov fdn I. QUES~ONTO~~~~OUI

This paper describes ongointg investigation into the subsumption control of robots. Subsumption architecture is a relatively new and simple approach to the control of robot systems. With subsumption control, the robiot is controlled by a set of simple processes. Each of the process is designed to provided the robot with a simple competence, for example, the ability to avoid contact with obstacles, to wander aimlessly around without hitting things or to explore the robot's environment. This paper includes: 1) reviewing subsumption architecture; 2) some methods proposedl to improve the emciency of this architecture; 3) the implementation of these methods and primitive abilities, such as, wandering behavior, collision avoidance, and navigation. The enhianced architecture is more effective than the original one.

In this paper, we propose a method for the visual-based location of a mobile robot with an omni-directional camera in indoor environment. Omni-directional camera is composed of conventional camera and convex mirror. It has large field of view, which further reduces the number of images necessary to represent a location. The stmctnre, p~operty anti geometry of the omni-directional camera are presented. The principle of trigonometric location is also presented. We use trigonometric location and data derived from omni-directional image processing to realize robot location. Preliminary location experiments with a mobile robot equipped with an omni-directional camera in indoor environment are performed. The result of experiments shows that the proposed method has good performance.