Intelligent Processing System Research Articles

Real-time segmentation of short videos under VR technology in dynamic scenes

Abstract This work addresses the challenges of scene segmentation and low segmentation accuracy in short videos by employing virtual reality (VR) technology alongside a 3D DenseNet model for real-time segmentation in dynamic scenes. First, this work extracted short videos by frame and removed redundant background information. Then, the volume rendering algorithm in VR technology was used to reconstruct short videos in dynamic scenes in 3D. It enriched the detailed information of short videos, and finally used the 3D DenseNet model for real-time segmentation of short videos in dynamic scenes, improving the accuracy of segmentation. The experiment compared the performance of High resolution network, Mask region based convolutional neural network, 3D U-Net, Efficient neural network models on the Densely annotation video segmentation dataset. The experimental results showed that the segmentation accuracy of the 3D DenseNet model has reached 99.03%, which was 15.11% higher than that of the ENet model. The precision rate reached 98.33%, and the average segmentation time reached 0.64 s, improving the segmentation accuracy and precision rate. It can adapt to various scene situations and has strong robustness. The significance of this research lies in its innovative approach in tackling these issues. By integrating VR technology with advanced deep learning models, we can achieve more precise segmentation of dynamic scenes in short videos, enabling real-time processing. This has significant practical implications for fields such as video editing, VR applications, and intelligent surveillance. Furthermore, the outcomes of this research contribute to advancing computer vision in video processing, providing valuable insights for the development of future intelligent video processing systems.

USE OF INTELLIGENT FUZZY IMAGE SEGMENTATION SYSTEMS IN THE PROFESSIONAL TRAINING OF FUTURE SPECIALISTS IN ENGINEERING AND PEDAGOGICAL FIELDS

Relevance: The article addresses the critical issue of integrating intelligent image segmentation systems that utilize fuzzy logic into the training processes for future specialists in engineering and pedagogical fields. This integration is a significant aspect of the digitization of higher education. Aim: The goal is to implement intelligent vehicle image segmentation systems using fuzzy logic to train specialists in engineering and pedagogical fields. Methods: The preliminary processing of the images of the studied objects (vehicles) involved digital filtering methods, contour detection, profile analysis, and contrast enhancement. Image segmentation was performed using watershed methods, contour lines, and region growing. After segmentation, the obtained segments were selected based on size. Fuzzy membership functions were then applied to determine the degree of affiliation of the segments to the meaningful parts of the studied objects, ensuring reliable recognition of these parts and stable operation of the intelligent system despite external influences on the acquired images. Results: A computer system has been developed for the segmentation of vehicle images using fuzzy logic, which has been integrated into the training of specialists in engineering and pedagogical fields. The segmentation methods isolate objects within the images, which are then recognized using fuzzy logic. Thanks to the fuzzy membership functions, elements of vehicle images are reliably recognized even when there is some ambiguity in the shapes of the segments. The practical significance of the developed system is demonstrated through the processing of car images. Conclusions: The integration of the developed system into the educational process provides students with both theoretical knowledge and practical skills related to intelligent image processing systems.

The Design and Execution of a Multimedia Information Intelligent Processing System Oriented to User Experience

The Design and Execution of a Multimedia Information Intelligent Processing System Oriented to User Experience

MOBILITY-AS-A-SERVICE: LITERATURE AND TOOLS REVIEW WITH A FOCUS ON PERSONALIZATION

In the coming years, mobility initiatives should focus on sustainability, safety, and social equity. This can be achieved by introducing innovative transportation methods, implementing novel approaches for end-users, and optimizing the utilization of traditional modes of transport. To achieve this goal, it is essential to utilize pervasive sensing and computing technologies, along with intelligent information processing systems, to assist decision makers, managers, and transport operators. To effectively address unforeseen events and disruptions, mobility services should promptly adapt and improve their flexibility. Furthermore, these services should be adaptable to meet the unique needs and evolving demands of individuals. Current research focuses on understanding how individuals make decisions about when and where they engage in walking, driving, and travel activities. Therefore, it is important to develop reliable human mobility models in this context. Big data and Artificial Intelligence (AI) are important in this context as they enable data generators to identify individual patterns and quickly adapt solutions. This paper aims to conduct a literature review on Mobility-as-a-Service (MaaS), focusing on personalization, to identify gaps in current MaaS initiatives. This assessment is essential for creating inclusive, user-friendly, personalized, and customizable MaaS solutions. To conclude, the existing challenges have been addressed in comprehending the characteristics of MaaS in terms of personalization. Additionally, they have been proposed further research questions to delve deeper into this aspect.

Revolutionizing Mushroom Identification: Improving efficiency with ultrasound-assisted frozen sample analysis and deep learning techniques

The aim of this research is to explore the potential applications of deep learning algorithms in categorizing ultrasound-assisted frozen mushrooms based on their quality attributes. To achieve this, an object detection model was trained on a dataset of images grouped by control (mushrooms frozen by immersion method without ultrasonication), pulse (mushrooms frozen by immersion method in combination with pulsed mode ultrasonication), continues (mushrooms frozen by immersion method in combination with continuous mode ultrasonication), and raw data (mushrooms before freezing. The accuracy of the developed model was 100 % for continues and control groups, while the model achieved accuracies of 92 % and 95 % for raw and pulse ones, respectively. Specifically, the model achieved a total accuracy rate of 96 % for identifying mushrooms. The model's excellent performance in accurately classifying images shows its ability to recognize subtle differences in image quality and clarity. The findings could provide valuable insights into using deep learning algorithms for analyzing mushroom quality data and developing intelligent mushroom processing systems.

Research and application of intelligent information processing system in power hierarchical control

Abstract In this paper, we propose a distributed secondary FM control method for microgrids to address the problem that the operating cost of microgrids increases significantly under load dropout and use the consistency algorithm to make the system reasonably distribute the load power to re-achieve the stable state. In the actual microgrid, the communication delay discusses the influence of information transmission and system stability. Simulation experiments are used to verify and set different convergence coefficients to obtain a suitable range of convergence coefficients. Then, the microgrid economic power allocation algorithm was used to describe the operating cost of microsources and limit the generator output power of the microgrid system to obtain the microgrid economic power allocation values. The cost amount of MG1 is $1327.3 in the equal micro-increase rate consistency calculation. The control condition test reduced 73.14 kWh in the overall load outage of intelligent microgrid control. The power system’s efficiency can be improved through a control method using intelligent information processing.

The Piezoionic Effect: Biomimetic Transduction Mechanism for Sensing, Actuation, Interface, and Energy Harvesting

AbstractThe transduction from mechanical to electrical signal or energy, and vice versa, has wide applications spanning wearable electronics, medical devices, human‐machine interfaces, prosthetics, synthetic biology, soft robotics, augmented reality (AR), internet‐of‐things (IoT), smart cities, environmental monitoring, energy conversation, and energy harvesting. Although many biological processes such as those in sensory receptors and cell communication uses ions, much of existing artificial intelligent sensory and processing systems operate based solely on electrons as signal or energy carriers. Therefore, the realization of natively ionic devices can significantly advance the frontiers of both capabilities and applications. This invited paper introduces the piezoionic effect, surveying the historical evolution, scientific description, methodology, application, future outlook, and current debates. This work complements existing ones in that it is an in‐depth analysis of key attributes of the piezoionics field but, unlike a conventional review, does not emphasize the categorization of individual works, though is also provided. This work is also aimed at facilitating the understanding and evaluation of the piezoionics field in terms of its key capabilities, promises, fundamental limitations, technical challenges, and practical applications. Piezoionics as a biomimetic mechanical‐electrical transduction mechanism is expected to grow rapidly. Through in‐depth analysis and working out implications for the future, it is hoped that this paper provides a foundation for further exploration, discussion, and collaboration across different fields. The community looks forward to the piezoionic effect evolving into a mainstream paradigm for the ubiquitous realization of sensors, actuators, interfaces, as well as energy harvesters and beyond.

Optimizing medical personnel speech recognition models using speech synthesis and reinforcement learning

Text-to-Speech synthesis (TTS) can be used to generate training data for building Automatic Speech Recognition models (ASR). Access to medical speech data is because it is sensitive data that is difficult to obtain for privacy reasons. Speech can be synthesized by mimicking different accents, dialects, and speaking styles in a medical language. Reinforcement Learning (RL), in the context of ASR, can be used to optimize a model. A model can be trained to minimize errors in speech-to-text transcription, especially for technical medical terminology. In this case, the “reward” to the RL model can be negatively proportional to the number of transcription errors. The paper presents a method and experimental study from which it is concluded that the combination of TTS and RL can enable the creation of a speech recognition model suited to the specific needs of medical personnel, helping to expand the training data and optimize the model to minimize transcription errors. The learning process used reward functions based on Mean Opinion Score (MOS), a subjective metric for assessing speech quality, and Word Error Rate (WER), which evaluates the quality of speech-to-text transcription. [The Polish National Center for Research and Development (NCBR) supported the project: “ADMEDVOICE- Adaptive intelligent speech processing system of medical personnel with the structuring of test results and support of therapeutic process,” no. INFOSTRATEG4/0003/2022.]

Synthesizing medical terms—Quality and naturalness of the deep Text-to-Speech algorithm

The main purpose of this study is to develop a deep text-to-speech (TTS) algorithm designated for an embedded system device. First, a critical literature review of state-of-the-art speech synthesis deep models is provided. The algorithm implementation covers both hardware and algorithmic solutions. The algorithm is designed for use with the Raspberry Pi 4 board. 80 synthesized sentences were prepared based on medical and everyday language employing the TTS algorithm developed. For tests, an application is built, containing a questionnaire allowing for evaluating the quality and naturalness of the synthesized speech, for both types of language. It is followed by the algorithm efficiency tests. A presentation of the performed tests, along with the results obtained from 30 respondents, is shown. The discussion consists of a statistical analysis of the obtained results and a comparison with other speech recognition solutions used as a reference. Finally, in the summary section, there is an overall conclusion of this approach and promising directions for future development. [This work is supported by the Polish National Center for Research and Development (NCBR) project: “ADMEDVOICE-Adaptive intelligent speech processing system of medical personnel with the structuring of test results and support of therapeutic process,” no. INFOSTRATEG4/0003/2022.]

ДОСЛІДЖЕННЯ ЕФЕКТИВНОСТІ ВИКОРИСТАННЯ РОЗПОДІЛЕНОЇ МАТРИЧНОЇ ФАКТОРИЗАЦІЇ В ПРОМИСЛОВИХ СИСТЕМАХ

The processing of big data is an exceedingly urgent challenge in the functioning of modern information systems. The latest information technologies must be employed to collect, store, and analyze vast amounts of information. Intelligent data processing systems were implemented in numerous fields, particularly in the industry. Smart industrial systems also utilize data from various devices, enabling automated management processes and network component analysis. A prime example of an intelligent industrial system is the smart grid, which efficiently distributes electricity to users by considering demand, network parameters, load, etc. Processing large amounts of information necessitates the use of machine learning methods and mathematical data analysis. Matrix factorization serves as an exemplary technique for transforming information into a more convenient form for further processing, establishing relationships between elements, and optimizing outcomes. In particular, the SVD (Singular Value Decomposition) and Funk-SVD algorithms are employed to address big data processing challenges, and they were discussed in this work. The key features of processing large data volumes in industrial smart grid systems were analyzed in the paper. The advantages of distributed computing for more efficient information analysis were identified. The recommendation algorithms that enable faster and more accurate processing of extensive data were explored in the study. Specifically, the SVD and Funk-SVD algorithms, used in recommendation systems for large data processing, were examined. A method of distributed matrix factorization to provide recommendations to smart grid system users was proposed in the paper. This approach involves the exchange of public data between devices and the local processing of private data. The advantages of this distributed model include flexibility in adjusting parameters, improved calculation accuracy through result exchange between nodes, high data processing speed, and scalability were identified. The conclusion that the proposed method can be effectively used in recommendation systems within the smart grid context, enhancing automated management processes and resource distribution was exclaimed.

Image Retrieval Through Free-Form Query using Intelligent Text Processing

Image Retrieval is the process of retrieving images from the image/multimedia databases. Retrieval of images are carried out with various types of queries, free-form query is a text-query that consists of single or multiple keywords and/or concepts or descriptions of images with or without the inclusion of wild-card characters and/or punctuations. This work aims to handle image retrieval based on free-form text queries. Simple & complex queries of conceptual descriptions of images are explored and an intelligent processing system with free-form queries based on the Bag-of-Words model is modified and built for natural scene images and on Diverse Social Images using the Damerau-Levenshtein edit distance measure. The efficacy of the proposed system is evaluated by testing 1500 free-form text queries and has resulted in a recall accuracy of 91.3% on natural scene images (of Wang/Corel database) and 100% on Diverse Social Images (of DIV400 dataset). These results show that the system proposed has produced satisfactory performance compared to published results such as the harmonic mean of precision and recall (i.e. F1-Score) of 76.70% & 63.32% at retrieval of 20 images etc in reported works.

An unsupervised linguistic-based model for automatic glossary term extraction from a single PDF textbook

Term extraction from textbooks is the cornerstone of many different intelligent natural language processing systems, especially those that support learners and educators in the education system. This paper proposes a novel unsupervised domain-independent model that automatically extracts relevant and domain-related key terms from a single PDF textbook, without relying on a statistical technique or external knowledge base. It only relies on the basic linguistic techniques of the natural language processing: pattern recognition, sentence tokenization, part-of-speech tagging, and chunking. The model takes a PDF textbook as an input and produces a list of key terms as an output. Furthermore, the model proposes a novel classification of sentences from which the concept of defining sentences is proposed. The defining sentences are the main textual units that the model revolves around to identify the key terms. The architecture of the proposed work consists of 21 processes distributed across three phases. The first phase consists of five processes for extracting text from a PDF textbook and cleaning it for the next phases. The second phase consists of eight processes for identifying the defining sentences and extracting them from all the textbook’s sentences. The last phase consists of eight processes for identifying and extracting the key terms from every defining sentence. The proposed work was evaluated by two experiments in which two PDF textbooks from different fields are used. The experimental evaluation showed that the results were promising.

Analysis and Research on Intelligent Management and Control System of Reusable Medical Devices

Focusing on the influencing factors of the operation and processing process of reusable medical devices, the management problems of reusable medical devices are analyzed from the processing processes of device assembly, packaging, handover, inventory and information recording. In the specific practice of designing the intelligent management and control system of reusable medical devices, the medical processes in different periods from device addition, packaging, disinfection, transfer, transportation, distribution, recycling to scrapping are integrated into the intelligent service system. Through the changes of medical device treatment, this study comprehensively explores the innovative ideas and specific problems in the construction of intelligent process system of hospital disinfection supply center.

Design of an Intelligent Processing System for Business Data Analysis Based on Improved Clustering Algorithm

In order to further optimize the processing of business data and increase the effectiveness and accuracy of business data analysis, this research investigated the design of an intelligent processing system for business data analysis based on improved clustering algorithms. Some issues with business data analysis can be efficiently resolved by enhancing clustering algorithms, resulting in the provision of more intelligent, thorough, and precise business data analysis services for businesses and organizations. The improved clustering algorithm is compared with the traditional clustering algorithm to analyze the experimental data, the results showed that the maximum clustering effect of the improved clustering algorithm was above 80%, the highest robustness was 91.65%, and the highest point in algorithm performance was 90.97%. From these three aspects, it can be seen that improved clustering algorithms have more advantages than traditional clustering algorithms. Therefore, improving clustering algorithms can make business data processing more efficient, accurate, and reliable, providing enterprises and organizations with higher quality business data analysis services.

Intelligent RGV Dynamic Scheduling Virtual Simulation Technology Based on Machine Learning

With the development of workshop automation, the complexity of RGV (Rail Guided Vehicle) dynamic scheduling schemes using virtual simulation technology is increasing. For the widely valued intelligent machining systems, machine learning based optimization algorithms can effectively respond to the increasingly complex RGV dynamic intelligent scheduling. In the whole model construction, how to complete the modular design of the intelligent processing system and optimize the solution is the key problem that needs to be solved urgently at present. This paper studied the use of particle swarm optimization to design the RGV dynamic scheduling model, aiming to improve the material processing production efficiency of RGV dynamic scheduling and reduce the system failure rate. Through problem modeling, solution and simulation experiment analysis, this paper applied particle swarm optimization based on machine learning, combined with RGV structure modular design and task parameter test set samples. According to the data results, the following conclusions can be drawn from the discussion. Under the background of intelligent logistics system, the RGV dynamic scheduling model using particle swarm optimization had higher material processing production efficiency than the traditional scheduling method in all job test samples, and the average increase was 13.25%. Meanwhile, in terms of system failures, optimization algorithms were better than traditional scheduling methods, with an average reduction of 4.6%. This shows that the RGV dynamic scheduling model based on particle swarm optimization has a better practical application effect.

DESIGN AND SYNTHESIS OF TERNARY LOGIC ELEMENTS

The aim of this paper is creating some ternary elements. The threshold element of ternary logic on bipolar transistors and elements of ternary systems based on it are considered. The main disadvantages of this approach are identified. The multi-threshold element of multivalued logic and its specific four-threshold implementation are considered. The scientific nowelty is the using of a multithreshold element of multivalued logic as a basis for constructing elements of ternary systems. It is shown that the advantages of a multi-threshold element of multi-valued logic are a larger number of thresholds of input signals, a larger number of levels that it can distinguish and a larger number of output signals. The implementation of some ternary elements, such as half-adder, disjunction and strong conjunction, is given. The practical significance of obtained results is that the multithreshold element of multivalued logic allows us to use it for synthesis basic elements of ternary logic and use one approach for all of them. Additional practical advantage is that in practice it allows you to build more diverse logical and arithmetic devices with a simplified implementation. The implementation of ternary devices based on threshold logic is a way to create ternary devices that can compete with binary devices in terms of equipment, capacity, operational capabilities, and variety. This, in turn, leads to greater speed and simplification of the structure of devices, as well as increasing the speed of data processing. These advantages are significant in such areas of computer use as intelligent data processing systems, expert systems, decision theory, i.e., where data analysis is performed. Prospects for further research are to use the method of construction and synthesis of nodes of ternary computer systems, based on MTEML. Their optimization, and development of principles of mathematical modeling and software of such systems and their elements is expedient.

Design and implementation of Pedestrian detection based on Zynq APSoc

Pedestrian recognition is an important content of intelligent security surveillance video processing system. In order to meet the application requirements of miniaturization and real-time security system, a moving target recognition system based on APSoc platform is designed. The system uses ov5640 camera as the video collector, zynq7020 as the development platform, and uses the soft and hard co-design to realize each function module. Finally, the pedestrian image acquisition and real-time display system is successfully completed. The test results show that the system can effectively and stably track pedestrians and moving objects within a certain range of distance, and can be displayed in real time, miniaturized and low power consumption, which can be further applied in the field of security.

Urban fire emergency management based on big data intelligent processing system and Internet of Things

At present, the necessity to obtain and collect logistics information data has also shown explosive growth. Because the past data management technology cannot meet the development requirements of IoT industry under new circumstances, intelligent processing technology based on big data has emerged in a historic moment. Using this technology, people can easily and quickly identify, retrieve, track, monitor and manage products. So we need to learn more about the IoT, to be able to properly handle various data, improve operational efficiency, maximize the value of the Internet of Things system, and accelerate social and economic development. With the acceleration of urbanization and the gradual improvement of the level of urbanization, the scale and number of cities have grown rapidly, and various high-rise buildings, large-scale assembly halls, and underground buildings continue to appear in the city. The increasing importance of this city is due to the accumulation of population and wealth, but it also faces unprecedented security issues. Because various buildings and assemblies are crowded with people, are used frequently, and have complex risk factors, fire accidents continue to occur. Therefore, emergency management of urban fires is indispensable. Among all kinds of fires, building fires are the most serious and direct danger to people. It is a matter that countries all over the world pay close attention to and monitor in real time.

Dadu-SV: Accelerate Stereo Vision Processing on NPU

Binocular vision and neural networks (CNNs) are widely seen in modern intelligent vision processing systems, such as robotics, autonomous vehicles, and AR gadgets. However, both the classic semiglobal matching (SGM) and deep CNNs entail substantial computing resource to reach the performance goal. Traditional embedded CPU/graphic processor unit (GPU) cannot simultaneously meet the processing speed and energy requirement, while the specialized circuits dedicated to SGM and CNN processing, respectively, will take considerable hardware and development costs. However, as the popularity of deep learning, neural processing units (NPUs) become prevalent in many embedded and edge devices, which possess high throughput computing power to deal with the matrix operations involved by neural networks. In this work, we attempt to take advantage of the neural processing architectures integrated in SoC chips to accelerate the SGM process, so that the hardware resources will be better utilized instead of investing more resources to create specialized SGM components. Thereby, this letter first deploys SGM on NPU by converting the incompatible operations into the neural-computing flow, and a configurable neural processing element is proposed to flexibly support various vector operation sequences. Then, a hybrid dataflow scheduler and the corresponding hardware modification are introduced to accelerate the cost processing, improving hardware utilization and on-chip memory footprint and access. Our solution runs at 45 fps for an image size of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$640\times 480$ </tex-math></inline-formula> , with 128 disparity levels. The speed-energy efficiency is <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$52\times $ </tex-math></inline-formula> better than the GPU (Jetson TX1) solution with negligible additional hardware overhead and accuracy loss.

A three-stage GAN model based on edge and color prediction for image outpainting

Human beings have a natural ability to perceive invisible things around them based on materials. This paper examines the problems encountered during image outpainting, which is widely utilized in computational photography, image editing and computer graphics. Although considerable progress has been made in image outpainting, problems such as semantic ambiguity, structural confusion and poor quality still arise. Inspired by the idea that humans draw a boundary contour, paint it with color, and then fill the content, we propose a three-stage GAN model, defined as the ECPIO network. The first-stage model is an edge prediction network (EP-net) that is used to predict the edge map of missing areas. The second-stage model is a color prediction network (CP-Net) that is utilized to predict the color map in the missing area, and the third-stage model is an image outpainting network (IO-Net) that is employed to generate the outpainting results. Since edge and semantics constrain the extended area, we introduce edge and semantic information to the discriminator to guide the image generation. Our method achieved a good performance with a PSNR of 27.15 and an SSIM of 0.78 for the Cityscapes dataset. The experimental results for other public datasets also show that our method can recover accurate semantic content and accurate structure information of missing regions. In addition, our method achieves good effects in semantic coherence, structure vision, color vision and texture vision. Our method has a wide range of intelligent image processing systems, such as computer photography, image editing, and image restoration.