FIGURE 9 - uploaded by Gerasimos Arvanitis
Content may be subject to copyright.
![Space filling curves [54].](profile/Gerasimos-Arvanitis/publication/344350000/figure/fig8/AS:938853158371341@1600851444375/Space-filling-curves-54.png)
Source publication
Nowadays, the preservation and maintenance of historical objects is the main priority in the area of the heritage culture. The new generation of 3D scanning devices and the new assets of technological improvements have created a fertile ground for developing tools that could facilitate challenging tasks which traditionally required a huge amount of...
Context in source publication
Context 1
... curve is superior to other curves in this respect [57]. Figure 9 depicts different space-filling curves as presented in [54]. In this work, two types of space-filling curves are investigated, simple reshape function, also referred to as sweep space-filling curve and Hilbert space-filling curve. ...
Citations
... We assume a small matrix as kernel, pass over the image, and compute feature map values according to (3), where the f indicates input image, and h indicates kernel. The m denotes rows and n denotes columns of the outcome matrix [30,[45][46][47][48]. ...
Nowadays, microscopy images are significant in medical research and clinical studies. However, storage and transmission of data such as microscopy images are challenging. Microscopy image compression is a vital area of digital microscope imaging in which image processing approaches are applied to capture the image by the microscope. It becomes accessible to interface the microscope to an image processing system because of technical advances in the microscope. Multiple application areas of microscope imaging, namely cancer research, drug testing, metallurgy, medicine, biological research, test-tube baby, etc., need microscopy image processing for analysis purposes. The microscopy image compression leads to complicated compression artifacts, like contouring, blocking, and ringing artifacts. Due to this problem, we select optimized Convolution Neural Networks (optimized-CNN), followed by Deep generative adversarial networks Deep-GAN, as a solution to reduce diverse compression artifacts. This research covers the compression of microscopy images and the removal of artifacts from a compressed microscopy image Optimized-CNN Deep-GAN based on Optimized-CNN and Deep-GAN. The concept of microscope image acquisition techniques and their analysis is also discussed. The performance of the Optimized-CNN Deep-GAN approach is measured using Peak Signal to Noise Ratio(PSNR), Compression Ratio(CR), Structural Similarity Index Measurement(SSIM), and Blind/Reference less Image Spatial Quality Evaluator(BRISQUE) and differentiated with state-of-the-art techniques. The experimental outcomes indicate the Optimized-CNN Deep-GAN technique acquires higher SSIM, BRISQUE, reduced space complexity, and better image quality than the existing image compression system. The proposed new model achieved CR 13.88, PSNR 40.6799 (dB), SSIM 0.9541, and BRISQUE 18.7645 values.
... Thus, structural configuration of an image is maintained as well. The model shown in Fig. 7 describes the CNN Forward as a well-designed network [38,39,40] which is used to represent the image in identified properties. Thus, CNN forward is used to compress in such a way that original data can be reproduced by reconstruction network. ...
... The optimum value is taken as maximum value which is done by Max pooling operation. Famous method called ReLU (Rectified Linear Unit) [32,36,37,38] activation function is used in CNN for a non-linear operation. This method activates neurons in which the gradient provides all times the optimum value. ...
... However, classification in predefined classes offers poor visual outcomes and limited flexibility in the case of vertex prioritization. Furthermore, Nousias et al. [28] presented a regression-oriented CNN architecture. The outcome was continuous saliency values with a greater level of detail. ...
Nowadays, 3D meshes are widely used in various applications in different areas (e.g., industry, education, entertainment and safety). 3D models are captured with multiple RGB-D sensors, and the sampled geometric manifolds are processed, compressed, simplified, stored, and transmitted to be reconstructed in a virtual space. These low-level processing applications require the accurate representation of the 3D models that can be achieved through saliency estimation mechanisms which identify specific areas of the 3D model representing surface patches of importance. Therefore, saliency maps guide the selection of feature locations facilitating the prioritization of 3D manifold segments and attributing to vertices more bits during compression or lower decimation probability during simplification since compression and simplification are counterparts of the same process. In this work, we present a novel deep saliency mapping approach applied to 3D meshes, emphasizing decreasing the execution time of the saliency map estimation, especially when compared with the corresponding time by other relevant approaches. Our method utilizes baseline 3D importance maps to train convolutional neural networks. Furthermore, we present applications that utilize the extracted saliency, namely feature-aware multiscale compression and simplification frameworks.
... Following this intuition and directives, current AI research and applications focus on the preventive aspects, and the improvement of the resilience of heritage. On this front, the EU project WARMEST 3 , proposed an AI method to monitor, inspect and assess potential deterioration on 3D digitised monument surfaces [32]. The deterioration may be the result of ageing, weathering or erosion and the proposed approach used deep learning to extract saliency maps and analyse surface structures to highlight potential regions of interest. ...
Recent advances in specialised equipment and computational methods had a significant impact in the Humanities and, particularly, cultural heritage and archaeology research. Nowadays, digital technology applications contribute in a daily basis to the recording, preservation, research and dissemination of cultural heritage. Digitisation is the defining practice that bridges science and technology with the Humanities, either in the tangible or in the intangible forms. The digital replicas support a wide range of studies and the opening of new horizons in the Humanities research. Furthermore, advances in artificial intelligence methods and their successful application in core technical domains opened up new possibilities to support Humanities research in particularly demanding and challenging tasks. This paper focuses on the forthcoming future of intelligent applications in archaeology and cultural heritage, by reviewing recent developments ranging from deep and reinforcement learning approaches to recommendation technologies in the extended reality domain.
... Basic Architecture of Convolutional Neural Network(Nousias S., et al. 2020).Algorithm2 describes the non-ROI compression of the medical image using CNN. We have trained and tested the CNN model using input images that pass through the network layers sequence with filters or kernels. ...
Medical image processing is an important field that directly impacts the health care system. It recognizes disease and also provides information for diagnosis and surgical process. The objectives of medical image compression are to reduce the computational complexity, storage size, and transmission bandwidth. This research has proposed an image compression scheme (MIC-DWT-CNN) based on discrete wavelet transform and convolutional neural networks. Region-growing and otsu-thresholding methods have separated the interested area and non-interested area of the medical image. The DWT has compressed the region of interest, and CNN has compressed the non-interested area in the medical image. The MIC-DWT-CNN scheme has experimented on the images of the medical image dataset using the python platform. The research objective is to achieve better compression efficiency and image quality. The performance of the MIC-DWT-CNN method has been evaluated using Mean square error (MSE), Peak Signal to Noise Ratio (PSNR), and Compression Ratio (CR). The existing techniques have been used to compare with the MIC-DWT-CNN method. The MIC-DWT-CNN method has achieved a better compression performance than the existing methods. The MIC-DWT-CNN method has achieved a higher CR, i.e., 25.01, than existing methods. Also, the model has provided the required level of MSE and PSNR values.
... State-of-the-art papers have investigated the use of semantic segmentation algorithms to automate the process of segmenting point clouds by architectural elements for 3D heritage documentation [32][33][34][35][36]. However, these methods were tested on large-scale photogrammetry surveys or laser scans, which may be impractical or prohibitively expensive for some heritage sites. ...
Crowdsourced images hold information could potentially be used to remotely monitor heritage sites, and reduce human and capital resources devoted to on-site inspections. This article proposes a combination of semantic image segmentation and photogrammetry to monitor changes in built heritage sites. In particular, this article focuses on segmenting potentially damaging plants from the surrounding stone masonry and other image elements. The method compares different backend models and two model architectures: (i) a one-stage model that segments seven classes within the image, and (ii) a two-stage model that uses the results from the first stage to refine a binary segmentation for the plant class. The final selected model can achieve an overall IoU of 66.9% for seven classes (54.6% for one-stage plant, 56.2% for two-stage plant). Further, the segmentation output is combined with photogrammetry to build a 3D segmented model to measure the area of biological growth. Lastly, the main findings from this paper are: (i) With the help of transfer learning and proper choice of model architecture, image segmentation can be easily applied to analyze crowdsourcing data. (ii) Photogrammetry can be combined with image segmentation to alleviate image distortions for monitoring purpose. (iii) Beyond the measurement of plant area, this method has the potential to be easily transferred into other tasks, such as monitoring cracks and erosion, or as a masking tool in the photogrammetry workflow.
... With the development of three-dimensional technology and the usage of VR & AR and Robotic systems, another field that is growing fast over the last years is 3D or multidimensional data. Finding points of interest in 3D clouds (Nousias et al., 2020a;Nousias et al., 2020b) or decomposing multidimensional workspaces into local primitives (Chamzas et al., 2019), becomes important and again corners (vertices) are one of them. An extension of Harris Corner Detection algorithm to 3D was proposed in (Głomb, 2009;Sipiran and Bustos, 2010;Sipiran and Bustos, 2011). ...
During the last years, the emerging field of Augmented & Virtual Reality (AR-VR) has seen tremendousgrowth. At the same time there is a trend to develop low cost high-quality AR systems where computing poweris in demand. Feature points are extensively used in these real-time frame-rate and 3D applications, thereforeefficient high-speed feature detectors are necessary. Corners are such special features and often are used as thefirst step in the marker alignment in Augmented Reality (AR). Corners are also used in image registration andrecognition, tracking, SLAM, robot path finding and 2D or 3D object detection and retrieval. Therefore thereis a large number of corner detection algorithms but most of them are too computationally intensive for use inreal-time applications of any complexity. Many times the border of the image is a convex polygon. For thisspecial, but quite common case, we have developed a specific algorithm, cMinMax. The proposed algorithmis faster, approximately by a factor of 5 compared to the widely used Harris Corner Detection algorithm. Inaddition is highly parallelizable. The algorithm is suitable for the fast registration of markers in augmentedreality systems and in applications where a computationally efficient real time feature detector is necessary.The algorithm can also be extended to N-dimensional polyhedrons.
Human attention is closely related to the visual information perceived. Saliency prediction models help to concentrate on the core pixels of an image contributing to attention. Trends and advancements in Artificial Intelligence resulted in many prediction models but not reached the level of prediction as human eye selection. Currently available prediction models concentrate on reflexive or exogenous attention. Published works in the year 2014–2022 are considered for this work and a study on improvements in the path made by deep learning prediction models using Convolutional Neural networks (CNN) in unveiling the pixels which achieve the high visual conspicuousness. This paper shows the avenues which are possible for the saliency models in the future by incorporating voluntary attention to the current models which are beneficial for researchers in the visual attention field like activity recognition, visual question answering etc.
Cyber-physical systems help toward mitigating the impact of climate change on cultural heritage sites and monuments by utilizing cultural heritage digital twins, coupled with artificial intelligence/machine learning algorithms, facilitating degradation detection and triggering appropriate actions.
In the recent decades, the significance of digital technology applications in archaeology and cultural heritage has been widely recognized, and digital applications have already contributed to the recording, preservation, study and dissemination of cultural heritage. This bridging, or even merging, begins with digitization, and particularly in three or more dimensions, when the objects of study are tangible. Once digital, the artifacts become available for a wide range of studies and dissemination activities. Recent advances in computing infrastructures and algorithmic techniques, empowered a re-ignition of artificial intelligence approaches, paving the way for a more data-driven future. On one hand, there is rich data production through modern multi-dimensional digitization and digital data production; on the other hand, there are powerful tools to effectively study and disseminate this rich content, ranging from artifacts’ analysis, to intelligent museums and personalized tourism. This chapter focuses on these two pillars towards a future of intelligent applications in archaeology and cultural heritage, by providing an account of modern state-of-the-art 3D digitization methods, as well as an account of AI applications, including deep and reinforcement learning, and automated recommendation systems for applications in the reality-virtuality continuum.KeywordsComputational archaeologyDigital humanitiesDigital archaeologyCyberarchaeologyCultural heritageArtificial intelligenceAI in archaeologyAI in cultural heritageDigitizationHeritage scienceHeritage analyticsExtended reality in cultural heritagePredictive modeling in archaeologyPersonalization in heritage applicationsCultural tourism
