Image classification is one of the important parts of digital image processing. We propose a novel feature space-based image classification method by combining manifold learning and mixture model. In this paper, the p...
详细信息
Image classification is one of the important parts of digital image processing. We propose a novel feature space-based image classification method by combining manifold learning and mixture model. In this paper, the process of image classification can be viewed as two parts: a coarse-grained classification and a fine-grained classification. In the coarse-grained classification, we apply the ISOMAP (Isometric Mapping) algorithm to do a dimensional reduction based on manifold learning. Thus, solving the classification problem is transformed from a high-dimensional data space to a low-dimensional feature space. And then, during the fine-grained classification, we present an improved EM algorithm of finite Gaussian mixture model to do clustering. Experimental results have demonstrated that the proposed method performs well in both accuracy and time. Additionally, our algorithm is robust to some extent.
With the increase of the number of college and university students,the amount of senate system data is correspondingly rising. As the core of university management,educational administration and its degree of informat...
详细信息
With the increase of the number of college and university students,the amount of senate system data is correspondingly rising. As the core of university management,educational administration and its degree of information are directly related to the university teaching quality and management efficiency. Compared with the traditional course enrolment,online course system can better reflect the wide application of information technology on campus,which was more convenient to communicate the information among schools,teachers,and students. Relying on the popular B/S three-tier architecture,we designed the online course system with a good interaction,scalability,and a strong security. This paper ***-based network technology elective system development process,outlined the function of online course selecting system,and its relevant technology.
This work proposes a novel three-layer federated learning (FL) framework with parameter selection and pre-synchronization (PSPFL) to achieve fast and accurate model training. The basic idea of PSPFL is that clients se...
This work proposes a novel three-layer federated learning (FL) framework with parameter selection and pre-synchronization (PSPFL) to achieve fast and accurate model training. The basic idea of PSPFL is that clients select partial model parameters for transmission and then base stations aggregate them cooperatively (i.e., pre-synchronization) and send the aggregated results to the server for global model update periodically. However, there is an intrinsic trade-off between parameter transmission overhead and model training loss. To strike a desirable balance between them, we investigate the optimal parameter pre-synchronization round and local training round under PSPFL. Specifically, we propose a Deep Q-Network (DQN)-based method to obtain the local training round and parameter pre-synchronization round. Finally, extensive experiments are conducted to evaluate the performance of the proposed method on commonly used datasets. The results show that the proposed method can reduce the sum of FL completion time and training loss by an average of 8.17%-18.82% compared to benchmarks.
The agriculture industry is one of the most significant sources of foreign exchange and employment in the Sri Lankan market. Therefore, small crops play a crucial role in ensuring the food security of the population a...
The agriculture industry is one of the most significant sources of foreign exchange and employment in the Sri Lankan market. Therefore, small crops play a crucial role in ensuring the food security of the population as they are integral components of Sri Lankan cuisine. However, industry experts have identified inefficient disease and pest management as a hindrance to production. This issue requires immediate attention and proper action, which can be effectively addressed through the utilization of ICT technologies. Consequently, this research paper proposes a smart system that aims to assist farmers and agriculture professionals by providing them with farmer education, disease and infestation management tools, as well as progression level calculations. The implementation of this system aims to uplift the agriculture industry in Sri Lanka. The developed ‘AI-based mobile application system for Brinjal Diseases,’ which is based on analyzing the eating patterns of leafhopper damage, presents an innovative approach to identify the most significant threats to Brinjal, specifically focusing on leafhopper damage. This proposed system effectively informs various stakeholders, including the Area Agricultural Research Officer, Seed Certification and Plant Protection Center (SCPPC), Agricultural Service Center, Office of the Register of Pesticides (ROP), Plant Protection Service (PPS), as well as farmers, regarding the dispersion and prevention of brinjal diseases and pest infections.
Portable Document Format (PDF), as one of the most popular file format, is especially useful for educational documents such as text books, articles, or papers in which we can preserve the original graphic appearance a...
详细信息
ISBN:
(数字)9781728181561
ISBN:
(纸本)9781728181578
Portable Document Format (PDF), as one of the most popular file format, is especially useful for educational documents such as text books, articles, or papers in which we can preserve the original graphic appearance and conveniently share online. Detecting and extracting information from tables in PDF files can provide a plethora of structural data to construct educational knowledge graphs. However, most of the existing methods rely on PDF parsing tools and natural language processing techniques, which generally require training samples and are frail in handling cross-page tables. In light of this, in this paper, we propose a novel OpenCV-based framework to extract the metadata and specific values from PDF tables. Specifically, we first highlight the visual outline of the tables. Then, we locate tables using horizontal and vertical lines and get the coordinates of tabular frames in each PDF page. Once the tables are successfully detected, for each table, we detect the cross-page scenarios and use the Optical Character Recognition (OCR) engine to extract the specific values in each table cell. Differing from other machine learning based methods, the proposed method can achieve table information extraction accurately without labeled data. We conduct extensive experiments on real-world PDF files. The results demonstrate that our approach can effectively deal with cross-page tables and only need 6.12 seconds on average to process a table.
An adaptive strategy based stable link selection algorithm is proposed in this paper, in which the stable neighbor metric and local movement metric are defined. On the basis, the stability probability of each link is ...
详细信息
An adaptive strategy based stable link selection algorithm is proposed in this paper, in which the stable neighbor metric and local movement metric are defined. On the basis, the stability probability of each link is computed adoptively to find the most stable link or route in a route discovery. The simulation results show that our algorithm can effectively adapt to the network conditions, and outperforms the longevity factor based algorithm as well as the residual lifetime based algorithm in selecting stable links.
The paper describes a project to develop an electronic map-based logistics service system (named as Eyou). It consists of several core components, including a unified portal for searching the goods delivery informatio...
详细信息
The paper describes a project to develop an electronic map-based logistics service system (named as Eyou). It consists of several core components, including a unified portal for searching the goods delivery information from different logistic firms, an electronic map-based interface by encapsulating Baidu map API, and the decision support component for planning the routes of the delivery vehicles in a feasible and cost-effective way. In this article, we describe the website's features and functions, the website's system architecture, a novel shortest path algorithm with time constraint and a case study based on our algorithm. The aim of Eyou is to provide some small-size logistics firms, the online shop owners and the users with three services in the convenient or inexpensive way.
We explore implementing a multilevel deep neural network to enhance the performance of a 4-channel photonic-electrical hybrid-packaged silicon transceiver. Stable transmission and reception of 150 Gbps/λ PAM4 signals...
详细信息
Our energy production increasingly depends on renewable energy sources, which impose new challenges for distributed and decentralized systems. One problem is that the availability of renewable energy sources such as w...
详细信息
An aggregate signature scheme allows a public algorithm to aggregate n signatures of n distinct messages from n signers into a single signature. By validating the single resulting signature, one can be convinced that ...
详细信息
An aggregate signature scheme allows a public algorithm to aggregate n signatures of n distinct messages from n signers into a single signature. By validating the single resulting signature, one can be convinced that the messages have been endorsed by all the signers. Certificateless aggregate signatures allow the signers to authenticate messages without suffering from the complex certificate management in the traditional public key cryptography or the key escrow problem in identity-based cryptography. In this paper, we present a new efficient certificate less aggregate signature scheme. Compared with up-to-date certificate less aggregate signatures, our scheme is equipped with a number of attracting features: (1) it is shown to be secure under the standard computational Diffie-Hellman assumption in the random oracle model, (2) the security is proven in the strongest security model so far, (3) the signers do not need to be synchronized, and (4) its performance is comparable to the most efficient up-to-date schemes. These features are desirable in a mobile networking and computing environment where the storage/computation capacity of the end devices are limited, and due to the wireless connection and distributed feature, the computing devices are easy to be attacked and hard to be synchronized.
暂无评论