Bayesian network (BN) is one of the most classical probabilistic graphical models. It has been widely used in many areas, such as artificial intelligence, pattern recognition, and image processing. Parameter learning in Bayesian network is a very important topic. In this study, six typical parameter learning algorithms were investigated. For the completeness of dataset, there are mainly two categories of methods for parameter estimation in BN: one is suitable to deal with the complete data, and another is for incomplete data. We mainly focused on two algorithms in the first category: maximum likelihood estimate, and Bayesian method; Expectation-Maximization algorithm, Robust Bayesian estimate,Monte-Carlo method, and Gaussian approximation method were discussed for the second category. In the experiment, all these algorithms were applied on a classic example to implement the inference of parameters. The simulating results reveal the inherent differences of these six methods and the effects of the inferred parameters of network on further probability calculation. This study provides insight into the parameter inference strategies of Bayesian network and their applications in different kinds of situations.

Advanced Intelligent Computing. Theories and Applications

Although a great deal of research effort has been devoted to the realistic and efficient rendering of terrain data, the existing research mainly focuses on displaying static terrain. The realistic simulation of dynamic terrain, including vehicle-terrain interactions, is absent in most 3D graphical applications. In this paper, we present a new GPU-based algorithm for dynamic terrain simulation. A key component of this algorithm is a Dynamically-Displaced Height Map (DDHM) which is generated and manipulated on the GPU. In addition, our method achieves real-time performance by using new features of the latest graphics hardware and shader technology. We demonstrate our algorithm by simulating a ground vehicle traveling on soft terrain. Our algorithm is particularly useful for ground based visual simulation applications as well as 3D games.

Real-Time GPU-Based Simulation of Dynamic Terrain

Microinjection of zebrafish larvae is widely used in vaccine development, drug screening, gene research, etc. Conventional manual injection has the disadvantages of low efficiency and operator-skill dependence. In this article, a high-throughput robotic microinjection system is proposed for zebrafish larvae. For the first time, the image contour-based potential energy algorithm is introduced to judge whether the microneedle has successfully pierced into the zebrafish larva, so as to further decide whether to inject materials into the target sample. The customized microstructured agarose medium can be used to fix batch zebrafish larvae with different poses in standard array for microinjection. A deep learning machine vision approach based on convolutional neural networks is employed to recognize multiple injection target points at one time. The multidevice collaboration achieves continuous and accurate microinjection operations. A prototype system has been fabricated for experimental testing. The results show that the system can quickly inject a batch of zebrafish larvae with a high success rate and high survival rate. Owing to a high degree of automation, the proposed microinjection system greatly reduces the workload of experimenters, saves the experimental cost, and shortens the relevant experimental study period. 

Design of a High-Throughput Robotic Batch Microinjection System for Zebrafish Larvae-Based on Image Potential Energy

Gray scale potential: A new feature for sparse image

A new method to recognize handwritten numerals is proposed.First,multi-classifiers is used to obtain some handwritten numerals' attributes to improve recognition accuracy.Then,by using rough set theory to reduce some attributes,it can improve the task's running time.Experiment results show that the new method is successful.

Recognition of Handwritten Numerals by Using Multi-classifiers Based on Rough Sets

A good feature extraction method can improve the performance of pattern recognition system or classification system. The contour can better to retain the original features of the object. In target recognition system , using potential energy of contour-point projection into the plane coordinate system. The method can be better to show a contour in the structural feature. In addition, it's better avoid the matrix storage redundancy. In all energy projection method, potential energy projection is better show its superiority in the structure information, the time of consumption and the storage space. Using potential energy theory into binary image feature extraction and feature store is a new method for image processing. Potential energy is to compute the energy of each pixel the first, and store the energy in order, which is better to retain the image features, and facilitate storage and classification. Contour is one of the most important features for distinguish between two objects. The contour potential energy can be used in target recognition and target classification field and so on.

Contour-based feature extraction for image potential energy theory

Due to the different imaging characteristics of sensor, there are big differences of multisource images in gray and trend of gray gradient. And the existing algorithms of image registration were time-consuming or low matching. In view of the status quo, a brief review of the SIFT algorithm is firstly given, and the shortcoming of SIFT, in which the matching rate is vulnerable to influence by gray feature, is pointed out. Then a fast algorithm for multisource image registration based on geometric feature of corners was presented. It adopts geometric feature of corners rather than gray feature. So the shortcoming of SIFT can be overcome. The novel algorithm can be used to register multisource images with large differences in gray or with different wavebands, and can increase the speed and raise the matching rate of registration. This section focused on how to select the corners, how to calculate feature vectors, and the feature matching algorithm. Finally, experiments have been done to prove that this algorithm can register images quickly and efficiently.

Fast algorithm for multisource image registration based on geometric feature of corners

Potential energy theory is a new method to extract target feature and recognize the target in image processing. Potential energy theory, when used for image processing, is faster, more efficient, and more accurate in target feature extraction. The data are easy to extract, and occupy small storage space, what’s more, they are simple to be computed with distinct features. Potential energy theory can also improve target recognition speed and recognition rate for the processing of target classification and target recognition. It is a totally new theory and method for image processing.

Potential Energy Theory Based on Image Processing

This paper presents a real-time GPU-based visualization of dynamic terrain in excavator simulator, which is an interactive system for the purpose of training human operators and corresponding researches. Firstly, we establish the kinematic model of excavator, so as to trans-fer the bucket tip's reference trajectories to corresponding, required reference angle sequence for each joint and to control motion sequences of articulated hierarchies. Then, our method use GPU to create the DDHM and digging offset map, with which we can perform the terrain de-formation entirely in the GPU. In order to improve the visual quality, we also present a method of procedural texture for deformation terrain. Finally, the Real-Time Optimally Adapting Meshes (ROAM) algorithm is used to render the dynamic terrain effectively.

Real-Time GPU-Based Visualization of Dynamic Terrain in Excavator Simulator

A good feature extraction method can improve the performance of pattern recognition system or classification system. Using potential energy theory into binary image feature extraction and feature store is a new method for image processing. The skeleton can be better display the whole features of the object. In target recognition system, using potential energy of skeleton-point projection into the plane coordinate system. The method can be better to show a skeleton in the structural feature. In addition, it can better avoid the matrix storage redundancy. In all energy projection method, potential energy projection is better shown its superiority in the structure information, the time of consumption and the storage space. The skeleton potential energy can be used in target recognition and target classification field and so on.

Skeleton-Based Feature Extraction Method for Two-Dimensional Potential Energy

Handle everyday research tasks with reliable, citation-backed results

Your personal Research Agent to handle research tasks with citation-backed results

Popular Tasks used by Researchers

How can I help with your research?

Meet SciSpace

Get more enhanced response by uploading the PDFs you want me to reference.

No relevant PDFs in your library

SciSpace is the AI research assistant for academics. Run systematic literature reviews on 280M+ papers, and write papers with cited sources. Trusted by 1M+ students, PhDs & researchers.

SciSpace | AI for Research

Analyze PDFs

Code & Manuscripts

Funding & Grants

Literature & Patents

Medical & Clinical Data

Systematic Review

Visualize & Present

Web & Data

Build a Google Scholar-like website for your research.

Build a website

Create charts and images for your research

Create a Chart

Write a paper for submission to a journal

Draft a manuscript

Patent Search

Design eye-catching scientific posters in minutes.

Scientific Poster Generation

Systematic Literature Review

One task is running at the moment. Your messages will be shown right after.

Drag and drop or click here to browse

Loved by <highlight>1 million+</highlight> researchers

Extract a list of specific topics and their sources from unstructured text

Topics

Compare and analyze relevant papers that matches with your search

Papers

Get insights from PDFs and bookmarked papers from your library

My library

Recent searches

Try searching for:

Catch AI-generated content in scholarly and non-scholarly content

{ai} Detector

Ai Writer

Get PDF Summaries, highlighted text explanations 

Chat with PDF

Effortlessly create in-text citations and bibliographies in APA and 2,500 other formats

Citation generator

Get explanations, summaries, and answers on academic papers

Ease up your research workflow with {scispace}'s cohort of exciting AI tools

Elevate your academic writing skills and convey your ideas the way you want

Paraphraser

Explore our range of reading and writing tools

Your file is being prepared and should be ready in a few minutes. If it's a large file, it might take a bit longer. You can close this window, and we'll email you the file when it's done.

You have reached a maximum limit of <strong>{limit}</strong> columns in the table. Remove at least <strong>1</strong> column to add or create another one.

Cheng Wang

Author Tools

Chat about Author