Ancient manuscripts written on both pages of the sheet are frequently affected by ink bleeding from the reverse side, which produces a significant degradation of the text. Effective digital image restoration techniques may require the use of the content of both document sides, thus needing their perfect alignment. Usually, recto and verso are not aligned either for rigid misalignments occurring during acquisition, or for non-rigid deformations of the sheet. In this paper we propose a novel method to jointly register and restore color recto-verso manuscript images in a piecewise manner, by subdividing the images into sub-images that exhibit apparent, different deformations of one with respect to the other. For each pair of corresponding sub-images, a specific projective transformation is computed, the two sub-images are registered, and then restored with a pixel-by-pixel algorithm that returns free of interferences versions of the images in their original acquisition layout. The projective transformation is estimated exploiting the precise computation of the shifts of a large number of small corresponding recto and verso patches, via correlation of their gradients. The experiments show that this combined procedure of local registration plus restoration can provide an excellent removal of bleed-through, while leaving unaltered the salient features of the original manuscripts.
In this paper we present a smart camera prototype capable of performing computer vison tasks directly on-board. The prototype is applied to real time monitoring of railways for detecting fast failures and other hazardous events to train circulation and providing notifications and early warnings to users. Experiments in a test site are reported together with encouraging preliminary results.
In the field of underwater cultural heritage, advanced technologies and tools are under development aiming at discovering, mapping, studying and securing archaeological sites. In this paper, we describe the system developed in order to disseminate the data collected during underwater exploration and to provide, to the archeologist, an effective tool to explore and analyze each explored underwater archaeological site. With this aim, the system exploits collected raw data together with the results of the detailed analysis and embeds them in a 3D interactive and informative scene. The scene is accessible both by the experts (for research purposes) and by general public (for dissemination of the underwater cultural heritage).
In the techniques proposed so far to remove bleed-through from digital images of ancient documents, two critical aspects are the identification of the occlusion areas, i.e. those pixels where the bleed-through pattern overlaps with the main foreground text, and the inpainting of the areas to be removed with a pattern that is in continuity with the surrounding background, often inhomogeneous due to paper texture or noise. In this paper we propose a new method for bleed-through removal that aims at solving both the aforementioned issues. The method first exploits information from the accurately registered images of the manuscript recto and verso to locate, in each side, the pixels corresponding to the interfering text, no matter if they are pure bleed-through or occlusion pixels. Then, processing separately the two sides, the identified areas are filled in by interpolating, through a suitable regularization model, the surrounding regions. We show the promising results obtained with this method on manuscripts affected by a very strong bleed-through.
We consider the problem of restoration of images corrupted by blur and noise. We find the minimum of the primal energy function, which has two terms. The former is related to faith fulness to the data and the latter is associated with smoothness constraints. In general, we have to estimate the discontinuities of the ideal image. We require that the obtained images are piecewise continuous and with thin edges. We associate with the primal energy function a dual energy function, which treats discontinuities implicitly. In order to have thin edges, we determine a dual energy function, which is convex and takes into account non-parallelism constraints. The proposed dual energy can be used as initial function in a GNC (Graduated Non-Convexity)-type algorithm, to obtain reconstructed images with Boolean discontinuities. In the experimental results, we show that the parallel lines are inhibited.
Underwater seafloors represent a huge archive of manmade artefacts, a cultural heritage whose abandon process started since man has been able to travel by the waters. In order to put this heritage under safeguard and preservation archaeologists require and request for technological support provided by the scientific community. Automated procedures tailored for the purposes of manmade object recognition in the underwater scenario represent the main topic discussed in this work. In particular the authors propose a set of procedures that enables to extract meaningful insights about the inspected environment and that can be exploited to assign a label of interest, in terms of cultural significance, to the surveyed areas. Furthermore the authors introduce a sketch of a framework according to which the identification of interesting areas on the seafloor may be implemented in terms of a Bayes decision system.
Recognition from partial views is a practical need for autonomous vehicles. Acquiring enough detailed information is not always possible. For avoiding obstacles we need partial recognition from a distance, before we have a full scanning in a dangerously close position. However, there are very few applicable results, especially in case of 3D. In this paper validation tests are presented for a method developed for partial recognition from 3D. For the real-life tests urban scenes of Mobile Laser Scanning data was selected. On the noisy LIDAR data, the excellent applicability of the method is proved. This paper shows the efficiency of the proposed method in real-life street scenery.
This paper investigates Kinect device application during rehabilitation of people with an ischemic stroke. There are many similar application using Kinect as a tool during rehabilitation. This paper is focused on measurement of Kinect's spatial accuracy and proposition of body states and exercises according to the Motor assessment scale for stroke (MAS). The system observes the whole rehabilitation process and objectively compares ranges of movement during each exercise. Angles between limbs are computed in the skeletal body joints projection to three anatomical planes, which enables a better insight to subject performance. The system is easily implemented with a consumer-grade computer and a low-cost Kinect device. Selected exercises are presented together with the angles evolution, body states recognition and the MAS Scale after the stroke classification.
This experimental study investigated possibility of face symmetry and symmetric face movements evaluation using MS Kinects HD face tracking. The main motivation for this research is facial paralysis or partial loss of muscle control caused by stroke. Evaluation of face symmetry can be used as an indicator of positive improvement of such disability. Precision of face model acquisition of MS Kinect v2 was estimated and two data sets of four facial exercises were recorded. Differences between data sets are evaluated with comparison of left part of face to the right one, with comparison of base frame with face at rest to frames during exercises, and with changes of angles of symmetric points on face. Results show significant differences between both sets, even though face tracking is affected by lightning, distance from camera and position angle of recorded person in view of sensor. Best results are achieved when comparing changes of angles (up to 6°) and differences of in symmetry of face (56% total symmetry in normal set to 13% symmetry in simulated set).
In this paper we address the problem of photogrammetric 3D reconstructions of an industrial cultural heritage site in Budapest, namely the Old Slaughterhouse. We perform an extensive comparison and evaluation of the state-of-the-art online visual SLAM (Simultaneous Localization and Mapping) and offline visual SFM (Structure from Motion) methods in order to obtain the 3D model of the building. We show results obtained using a dataset recorded with a camera-equipped Micro Air Vehicle.
Autonomous underwater vehicles (AUVs) have been used mainly in naval operations and, in lesser extent, in marine biology and underwater archaeology. Before the ARROWS project, no AUVs specifically designed for archaeological research existed. In the framework of ARROWS, three AUVs were developed with the capabilities to map the seabed, study the discovered objects and penetrate the shipwrecks. In the summer of 2015, two of those AUVs - MARTA and U-CAT were field-tested in Estonia along with two commercially available AUVs. This paper presents the results of the tests.
Crime surveys are conducted to record crimes by the Office for National Statistics (ONS) in the United Kingdom every year. They contain rich information about crime. They record the crimes that are not reported to the police. However, their exploitation for gaining a better understanding of crime activities is limited. When used, traditional statistical models and descriptive statistics are adopted. The data they contain is very complex and changes from one year to the other. In this paper, we report the preprocessing activities that were performed on survey data to allow their use with data mining models. We reported the results of early analysis of the survey data using decision trees and the users' interpretation of these results.
We present a strategy for transparent, robust watermarking to protect intellectual property rights on cultural heritage images. This can be applied on any kind of vector image, from color to hyperspectral, and has a particular value when each individual channel has its own significance and can be used independently of the others, as often happens with quantitative diagnostic images. For color images, we can reduce the correlation between channels by relying on alternative color spaces. Dealing with images with any number of channels, we rather propose to work in the principal component space, whatever the watermarking algorithm chosen. We motivate why this is advantageous and show an example experiment using an embedding procedure proposed in the literature.
With development of new technologies, many applications generate large volumes of data that all need to be collected and processed instantly. Flowing as streams, these data are usually continuous, voluminous and cannot be stored integrally as persistent data. In this context, new systems called Data Stream Management Systems (DSMS) have emerged for processing data streams on the fly. However, in some applications, we can analyse expired data. Treating a data stream is performed according to a well defined temporal window. Beyond this window, data are discarded or lost forever. Some applications need to keep track of expired data. Thus, it is necessary to retain a compact structure (synopsis or summary) of streams in order to answer a wide range of needs. In this paper, we are interested in developing a generic summary structure for expired data. In order to preserve the possibility of performing future analysis, we suggest to establish specifications on these expired data. These specifications called forgetting functions define summaries (by aggregation) to be retained among the data to `forget'. We apply our approach to a real dataset for building summaries. A data cube is set up to answer a variety of needs.
General methods of video processing and three dimensional modelling have a wide range of applications in engineering, archaeology and spacial objects study. The paper is devoted to applications of these methods in biomedicine and neurology using MS Kinect depth sensor for non-contact monitoring of breathing. A special attention is paid to visualization of results and motion mapping over the selected chest area. The proposed methodology applies digital signal processing methods and functional transforms for acquired data de-noising, spectral analysis, and feature selection. Suggested method uses further the local polynomial approximation to detect extremal values of spectral components. The results verify the correspondence between the evaluations of the breathing frequency obtained from the thorax movement recorded by the depth sensor. The study proves that simple depth sensors can be used for non-contact detection of breathing frequency and for the three dimensional modelling of the chest movement. The proposed non-contact method enables to analyse breathing for diagnostic purposes and monitoring in the home environment as a component of assisted living technologies. General methodology studied form a contribution to the use of video sequences or sets of images for spacial objects modelling, their recognition, possible three dimensional printing or analysis of time evolution of their features.
This paper proposes a vessel recognition and classification system based on vessel acoustic signatures. Teager Energy Operator (TEO) based Mel Frequency Cepstral Coefficients (MFCC) are used for the first time in Underwater Acoustic Signal Recognition (UASR) to identify platforms the acoustic noise they generate. TEO based MFCC (TEO-MFCC), being more robust in noisy conditions than conventional MFCC, provides a better estimation platform energy. Conventionally, acoustic noise is recognized by sonar operators who listen to audio signals received by ship sonars. The aim of this work is to replace this conventional human-based recognition system with a TEO-MFCC features-based classification system. TEO is applied to short-time Fourier transform (STFT) of acoustic signal frames and Mel-scale filter bank is used to obtain Mel Teager-energy spectrum. The feature vector is constructed by discrete cosine transform (DCT) of logarithmic Mel Teager-energy spectrum. Obtained spectrum is transformed into cepstral coefficients that are labeled as TEO-MFCC. This analysis and implementation are carried out with datasets of 24 different noise recordings that belong to 10 separate classes of vessels. These datasets are partially provided by National Park Service (NPS). Artificial Neural Networks (ANN) are used as a classification method. Experimental results demonstrate that TEO-MFCC achieves 99.5% accuracy in classification of vessel noises.
To cope with heterogeneity of data in streams, Semantic Web technologies (RDFVSPARQL 2 ) have recently been used for annotation, publication and reasoning on these data. To deal with this new kind of streams, researchers have proposed new systems named RDF Stream Processing (RSP). Unfortunately, in limited system resources environment, these systems are fallible as soon as their maximum supported speed is reached. To overcome these problems, some efforts have been done in this area. Most of them, based on a triple-oriented approach and according to a probabilistic method, decrease the volume of RDF data stream using load-shedding techniques. In this paper we propose an enhancement of a Graph-Oriented approach for load-shedding semantic data streams, by considering the continuous query as input. Conducted experiments show that we can keep the RSP's recall at 100% even if we drop more than half of data.
The paper presents an outcome of the VISAS project (www.visas-project.eu) that concerns a virtual reality application for the exploitation of the underwater cultural heritage. The VR system takes advantage of novel 3D reconstruction techniques to provide geolocated and multi-resolution textured 3D models of underwater archaeological sites. Within the virtual underwater sites users live a recreational and educational experience by receiving historical, archaeological and biological information and contents about the submerged exhibits and structure of the site. Furthermore, the VR system allows divers to make a detailed planning of the operations and itinerary that will be later performed in the underwater environment.
Museums and other cultural heritage custodians are interested in digitizing their collections, not only for the sake of preserving cultural heritage, but also to make the information content accessible and affordable to researchers and the general public. Once an objects digital model is created it can be digitally reconstructed to its original uneroded or unbroken shape or realistically visualized using different historical materials. Some artifacts are so fragile that they cannot leave the carefully controlled light, humidity, and temperature of their storage facilities, thus they are already inaccessible to the public, and the viable alternative is their exhibition in the form of an augmented reality scene. We present a sophisticated measurement and processing setup, which we have developed, to enable the construction of physically correct virtual models. This setup is illustrated on the reconstruction of one of the best known Celtic artifact from the European Iron Age period to its original uncorrupted form.
The paper presents a new method for separation of overlapping dental objects allowing more precise analysis of the dental arch using image processing methods and computational intelligence. The methodological part of the paper is devoted to the separation of dental objects using the watershed and region growing methods applied to denoised images of dental bodies. The main part of the paper is devoted to analysis of image components and to the use of normal vectors to image segments boundaries to detect separate dental arch components in the complex environment of overlapping bodies. Experimental part of the paper is devoted to the application of the proposed method to the processing of real images of dental plaster casts acquired with the different illumination. Results include the proposed general algorithm for image components separation allowing further evaluation of dental arch parameters during the operation to allow the efficient treatment in the clinical environment.
