人文学研究における原資料デジタル画像の利用と認識

With the rapid development of digital image processing and recognition, the intelligent grid wins a great opportunity. We could be able to get the information from digital images quickly based on computer analysis, algorithms of deep learning and machine learning, grid data extraction, model training and image recognition. The article firstly reviewed the current status of traditional grid, including a broad overview of the specific strengths and weaknesses, secondly analyzed the intelligence algorithms of deep learning and machine learning based on neural network, then achieved the recognition and reconstruction of wiring diagram through image recognition based on neural network, lastly showed the prospects of image recognition in power system.

Digital image recognition based on Fractional-order-PCA-SVM coupling algorithm

Image processing and recognition methods are useful in many fields. According to situation, different techniques are used. For many years, methods based on optical Fourier transformation were very popular. Image recognition was performed generally by using optical correlators. Correlation techniques were strongly developed especially for military applications, but in many cases (industrial, biological and biomedical applications) these techniques suffer from a number of limitations. For these reasons, methods based on extraction and statistical processing of image features are more useful. Set of features can be extracted directly from an image (features based on image morphology, image moments etc.) or from image transforms (Fourier, Radon, Hough, Sine, Cosine etc.). The Fourier transformation is one of the most important in image processing. It can be simply performed by using an optical diffractometer. It allows to build image descriptors independent on image translation and after processing independent on image rotation. Diffractometers are very convenient in industrial and medical applications. Digital image processing and recognition were strongly developed on powerful workstations, however these procedures can also be implemented in PCs with DSP microprocessor cards or in situations where digital transforms used for image processing can be simply implemented and do not consume a lot of time. The example of biomedical image recognition performed in an optical way, by using a diffractometer, and in a digital system with a CCD camera will be described here.

단순히 인문학 자료를 디지털화 하거나, 연구 결과물을 디지털 형태로 간행하기보다는, 정보 기술의 환경에서 보다 창조적인 인문학 활동을 전개하는 것을 ‘디지털 인문학’이라고한다. 무엇이 ‘디지털 인문학적 연구’인가 라는 물음에 답을 줄 만한 연구 모델은 아직 명확하게 정립되어 있지 않다. 하지만, 최근 수년 사이 ‘디지털 인문학 연구’로 보고된 여러형태의 인문 분야 연구에서 가장 두드러진 공통점을 찾는다면, 그것은 종래에 인간의 언어로만 기술되었던 인문 지식의 요소들을 기계가독적(machine readable) 데이터로 전환하여, 컴퓨터가 그 정보의 해석을 도울 수 있게 하는 것이라고 할 수 있다. 필자는 이 글에서, ‘무엇이 디지털 인문학인가’하는 원론적 논의에서 한 걸음 더 나아가 인문 지식을 데이터로 전환하는 것은 어떠한 작업이며, 그것을 통해 학술적 성과를 기대 할 수 있는지를 설명하고자 한다. 이를 위해 한국학중앙연구원 장서각과 인문정보학 교실에서 수행하고 있는 데이터 기반 문중고문서 연구를 사례로 제시하였다. 디지털 인문학의 프레임워크 위에서 만들어진 새로운 형식의 고문서 디지털 아카이브는 연구자들이 직접 그 디지털화된 문서의 내용을 데이터로 취급할 수 있게 한다. 그리고 데이터의 분석, 정리, 해석을 통해 찾아낸 ‘사실’과 ‘이야기’를 확장된 데이터로 축적해 나아갈 수 있게 함으로써, ‘자료’의 아카이브에서 ‘지식’의 아카이브로의 진화를 지향한다. 고문서 내용을 데이터로 전환하여 기계적 가독성을 부여하는 것은 일차적으로 과거의 생활사에 대한 인문학적 연구에 기여하겠지만, 향후 이러한 데이터의 광범위한 축적으로 만들어질 인문학 빅데이터는 현대에 발생하는 사회과학적, 산업적 빅데이터와 엮여져서 현대사회에 대한 이해와 과거 역사의 탐구 사이의 연결고리를 만들어내는 것도 가능하게할 것이다. 인문지식이 데이터로 소통될 수 있도록 하는 노력을 통해 우리의 인문지식은 전통적인 인문학의 경계를 넘어서는 세계와도 교섭할 것이며, 인간과 사회에 대한 이해로서의 인문학의 외연을 더욱 의미 있게 확장해 갈 수 있을 것이다.Rather than simply digitizing humanities materials or providing humanities research results in a digital format, “digital humanities” encompasses ever more creative humanities endeavors in an information technology environment. There is still no research model which can clearly answer the question of what “digital humanities research” looks like. However, if we look at the similarities among humanities research which has been deemed to fall under the umbrella of “digital humanities research,” we can see the transformation of humanities knowledge elements from human-composed language to machine-readable data, which allows computers to aid in the interpretation of said information. The author will go a step beyond the fundamental discussion of “what digital humanities is,” to explain the nature of the work of transforming humanities knowledge into data, and what kind of scholarly outputs we can expect from such work. To this end, the author presents research led by the Academy of Korean Studies’ Jangseogak Archives and Cultural Informatics Lab which is based on old documents from paternalistic lineage organizations (munjung). A new form of digital archive of old documents made for a digital humanities framework allows researchers to turn the content of digitized old documents into data. In addition, through the accumulation of data which encompasses “truths” and “stories” discovered via data analysis, organization, and interpretation, an archive of “source materials” can evolve into an archive of “knowledge.” On preliminary level, transforming the content of old documents into data and making it machine readable contributes to humanities research on the lives of people of the past. However, looking forward, the widespread accumulation of such data can create big data on the humanities. And if this humanities big data is connected to big data dealing with contemporary social, academic, and business phenomena, it can serve as a link between the understanding of modern society and the history of the past. Through efforts to allow humanities knowledge to communicate as data, our humanities knowledge can go beyond the boundaries of traditional humanities and negotiate with the larger world, and it can more meaningfully expand the full realm of the humanities as an understanding of humans and society.

Based on chaos synchronization of two response lasers (RLs) injected by the common chaos signals from a driving laser (DL) with double optical feedback (DOF) and an additional neural network (NN), a bidirectional chaotic secure communication system with machine learning capability is proposed and the long-distance digital image transmission and recognition performances are numerically investigated. The simulated results show that the time delay signature (TDS) of the driving chaotic signals from DL with DOF is greatly weakened under suitable operation conditions. Two RLs can generate chaos signals with low TDS and broad bandwidth, and the high-quality chaos synchronization between two RLs can be achieved while the synchronization quality between DL and any one of RLs is extremely bad, which is helpful to realize the secure communication. Using the synchronized chaos output signals from two RLs as chaos carriers and chaos masking (CM) encoding method, the bidirectional 10 Gb/s secure communication over 140 km fiber channel can be realized and the effects of the fiber parameters on the communication performances are discussed. After adopting an additional NN, a long-distance digital image transmission and recognition based on this proposed secure communication system can be realized and the communication distance can be extended to 220 km.

Background: With the rapid development of digital twinning technology, the compatibility of digital twinning technology to other technologies is continuously enhanced. It is because of this that the application of image recognition technology in digital twinning technology becomes a reality. However, the key technology of digital twin, the virtual-actual mutual control technology, is not mature enough, and the image recognition technology applied in digital twin technology also has the problem of coordinate system transformation, which becomes an important link of image recognition technology in digital twin. Methods: Based on the above two problems, we take the tangram splicing project as an example to realize a virtual-actual mutual control method, so that the digital twinning technology can be well presented. Furthermore, we implement an image recognition applied to the conversion of digital twinning technology, so that digital twinning technology and image recognition have a seamless connection, allowing the application range of digital twinning technology to be further expanded. Results: In this paper, image recognition technology is successfully applied to digital twin technology by adopting the conversion between different coordinate systems and the real and virtual real time control, which makes the applicability of digital twin technology in high-tech fields such as smart factories and smart manufacturing to a higher level. Conclusions: Finally, through the tangram splicing project, the motion trajectories of the robotic arm and the tangram are consistent with the virtual robotic arm and the tangram in the computer. It is proved that our method can well combine digital twinning technology and image recognition technology.

Digital filtering is essential for digital imaging, image recognition, and super-resolution technology. For example, the presence of noise in images captured by digital cameras causes deterioration of the image quality and image recognition rate. In order to improve the image recognition rate, noise reduction and edge preservation must be performed during preprocessing. Noise is generally reduced using low-pass filters, such as the Gaussian filter. Although they reduce noise, such filters also have the properties of blurring edge. A strong edge blur reduces the accuracy of the feature detection in image recognition. Therefore, in our previous study, a fast M-estimation Gaussian filter for images (FMGFI) was proposed as an image filter that simultaneously achieves denoising and edge preservation. In the FMGFI, the setting of the optimal basic width of the 2nd order B-spline basis functions is important for achieving simultaneous denoising and edge preservation. In this method, the optimal basic width of the FMGFI was determined not only by manually setting the basic width but also by human judgment of the filtered images. Consequently, the inability to automatically determine the optimal basic width hindered efficient denoising during image processing Therefore, in this research, we develop and propose a method that can automatically determine the optimal basic width of the FMGFI. The previously proposed method calculates using the same basic width for all the pixels over the entire image; in contrast, the proposed method calculates using the basic width automatically determined for each pixel. The experiments confirmed that the method proposed in this study achieves higher denoising and edge preservation performance than the ones used in previous research. The results also showed that it has the highest denoising performance against salt-and-pepper noise as compared to other filters: non-local mean filter, Gaussian filter, median filter, bilateral filter, adaptive bilateral filter, and FMGFI. The experimental results for the Gaussian noise sowed that the proposed method has the same denoising and edge preservation performance as the other filters in visual evaluation. From the above, the proposed method is expected to contribute to efficient denoising and improvement of image quality by using it as a preprocessing.

With the exponential improvement of the integration capabilities of electronic hardware devices, digital images have become an indispensable information carrier, thus promoting the development of image recognition, detection, or tracking technologies. At present, there is a blank in the field of image recognition measured by the action standard of yoga standing three-dimensional movement, which will bring huge room for researchers to play. Based on the traditional image recognition technology, there are cases where missed detection or false detection cannot be correctly identified and made a correct judgment. In this paper, the ability to capture the action is improved by optimizing the network for the positioning of the stereoscopic action target of the yoga station. The linear discriminant analysis (LDA) after adding optimization is used to reduce the dimensionality of the captured image data, which is beneficial to improving image recognition rate and reducing image loss rate. Through the analysis of the three-dimensional movement of the yoga station, the model based on the improved algorithm is compared with the traditional model, and the final image recognition accuracy is improved compared with the network model before the improvement. The image recognition error rate steadily tends to 5%, and the loss rate is also below 2%. Through the optimized convolutional neural network, this paper can accurately capture the image position, and the recognition rate has also been greatly improved, which can provide a reference for future research in other fields.

This paper investigates a fractionation method of cellulosic fibres in dry state. Although processing of cellulosic fibres in airstream has been utilised for decades in airlaid processes, no attempts to fractionate fibres in dry state have been reported. Our main goal is to determine how fibre properties and fibre consistency affect fibre flocculation and thus the behaviour of fibres in fractionation. Also evaluations of quality and capacity of fractionation operation are made. Pulp flocculation behaviour is measured with digital imaging and image recognition technique before and after the fractionation device. A doubleview, orthogonal imaging approach is applied obtaining statistics of three-dimensional shape, dimensions and coordinates of detected fibre flocs. Fraction and fibre properties are determined with Metso Fractionator and Metso Fiberlab. Results show that fibre behaviour in a given system strongly depends on the pulp properties. Flocculation was found to be strongly dependent on fibre consistency. Evaluation of fractionation operation was done with help of calculations of mass-reject rates and separation efficiencies. Calculations showed that highest separation efficiencies can be achieved with system when mass-reject rate is between 0.30 and 0.40.

This paper explores teenage girls' engagement with digital images on social media. Using new feminist materialism, we foreground digital images as an assemblage of materialities (human and more‐than‐human) filled with affective potentials that materialise in/capacities. Drawing from interviews and focus group discussions, we show how the production and sharing of selfies through posting and sexting unlocked new ‘becomings’ through expressions of heterosexual desirability and pleasure but also generated fear through sexual objectification, sexual double standards and harassment. A recognition of digital images as materially embodied through which unequal gender power relations materialise is vital to addressing online sexual risk.

We present a classification algorithm based on ant colony system(ACS) for classifying digital images. The ACS has been recently emerged as a useful tool for the pattern recognition, image extraction, and edge detection. The classification algorithm of digital images is very important in the application areas of digital image coding, image analysis, and image recognition because it significantly influences the quality of images. The conventional procedures usually classify digital images with the fixed value for the associated parameters and it requires postprocessing. However, the proposed algorithm utilizing randomness of ants yields the stable and enhanced images even for processing the rapidly changing images. It is also expected that, due to this stability and flexibility of the present procedure, the digital images are stably classified for processing images with various noises and error signals arising from processing of the drastically fast moving images could be automatically compensated and minimized.

As the application of electronic equipment in people's production and life is becoming more and more common, digital images have become an indispensable means of information, and massive amounts of image data are being generated every moment. Image recognition is a very important research field in the field of computer vision, and the accurate recognition of objects in images is becoming more and more valuable. The purpose of this paper is to study deep neural network image recognition based on an improved LSTM. The five methods and steps of image recognition, the classification of convolutional neural networks on event images, the method based on LSTM subdivision and the existing event image classification methods are introduced. The experimental part introduces the parameter settings, and gives each visual the experimental results and analysis under the characteristics. Through the comparison of classification accuracy, after LSTM network processing, the event with the highest accuracy rate is 98%, which proves the superiority of the network structure constructed in this paper in the problem of event image recognition, and the classification performance has been improved to a certain extent.

This paper aims to propose deep learning-based approach to the auto-recognition of their design features of interior design elements using given digital images.The recently image recognition technique using convolutional neural networks has shown great success in the various field of research and industry.The open-source frameworks and pre-trained image recognition models supporting image recognition task enable us to easily retrain the models to apply them on any domain.This paper describes how to apply such techniques on interior design process and depicts some demonstration results in that approaches.Furniture that is one of the most common interior design elements has sub-feature including implicit design features, such as style, shape, function as well as explicit properties, such as component, materials, and size.This paper shows to retrain the model to extract some of the features for efficiently managing and utilizing such design information.The target element is chair and the target design features are limited to functional features, materials, seating capacity and design style.Total 3933 chair images dataset and 6 retrained image recognition models were utilized for retraining.Through the combination of those multiple models, inference demonstration also has been described.

Kuzushiji, as characters written in a cursive style, had been continuously used for both publishing and handwriting in Japan before the end of the 19th century. However, well into the twentieth century, most modern Japanese lost the ability to read Kuzushiji due to changes in writing systems. Therefore, how to develop advanced machine learning algorithms to identify Japanese historical characters has great significance in preserving cultural knowledge and promoting sustainable development in urban living. Unlike traditional digit image recognition, Kuzushiji image recognition is a more challenging task due to the complex script for cursive handwriting, multiple variations between characters, and imbalanced data. To address such issues, a novel deep learning‐based character recognition method called residual shrinkage balanced network (RSBN) is developed for classifying Japanese historical characters. A novel residual shrinkage structure with a soft threshold function is constructed to reduce redundant features using specially designed skip‐connected subnetworks. Then, multiple residual shrinkage modules are stacked together to learn hierarchical features from raw input. Furthermore, a novel class rebalancing strategy is further designed to improve learning performance from imbalanced data. The experiments are conducted on Japanese classical literature datasets. The results demonstrate that the proposed method obtains more than 89.01% testing accuracy on Kuzushiji‐MNIST dataset, which outperforms the existing deep learning‐based character image recognition methods. Moreover, without additional data augmentation techniques, the proposed approach also achieves 97.16% balanced accuracy on Kuzushiji‐49 dataset, which is the best one compared with existing published results. When considering the computational performance, the number of trainable parameters of the proposed method is competitive with that of the typical ResNet‐18.

The visual representation of emotion in image technology is particularly important in today's society. Visual perception, which is mainly based on the changes in the audience's psychological emotions, has become a thinking derivative of digital image technology, and further technical implementation of image transformation, adjustment, repair, reconstruction, replacement, etc. Ingesting the characteristics of emotional appeal is the main direction of this research. With the continuous development of blockchain, virtual and information technology, the research on the content and emotion of digital images has become a hot spot in the design of visual reconstruction. A multi-dimensional computer-generated language spanning space is proposed to improve the methods of digital image acquisition, recognition, transformation and segmentation. Reconstruction design research on the semantic characteristics of computer-generated digital images and content emotional representations. Based on the multi-features of digital images, the differences in image shape, texture, color, motion and spatial domain features are analyzed, and the content emotion and semantic reconstruction of digital images are distinguished. In information transmission, multi-dimensional space capture, output, and quantitative analysis methods are used to establish a new sample arrangement matrix of digital images, select the best parameters for content expression, and reconstruct and process quasi-vision, combined with parameters to realize emotion recognition of digital images. The detection coefficient and dimension are determined by simulation and virtual technology, and the high recognition rate of digital image is obtained, and a good digital image technology display effect is achieved.