Search

Geohazard recognition and inventory mapping are absolutely the keys to the establishment of reliable susceptibility and hazard maps. However, it has been challenging to implement geohazards recognition and inventory mapping in mountainous areas with complex topography and vegetation cover. Progress in the light detection and ranging (LiDAR) technology provides a new possibility for geohazard recognition in such areas. Specifically, this study aims to evaluate the performances of the LiDAR technology in recognizing geohazard in the mountainous areas of Southwest China through visually analyzing airborne LiDAR DEM derivatives. Quasi-3D relief image maps are generated based on the sky-view factor (SVF), which makes it feasible to interpret precisely the features of geohazard. A total of 146 geohazards are remotely mapped in the entire 135 km2 study area in Danba County, Southwest China, and classified as landslide, rock fall, debris flow based on morphologic characteristics interpreted from SVF visualization maps. Field validation indicate the success rate of LiDAR-derived DEM in recognition and mapping geohazard with higher precision and accuracy. These mapped geohazards lie along both sides of the river, and their spatial distributions are related highly to human engineering activities, such as road excavation and slope cutting. The minimum geohazard that can be recognized in the 0.5 m resolution DEM is about 900 m2. Meanwhile, the SVF visualization method is demonstrated to be a great alternative to the classical hillshaded DEM method when it comes to the determination of geomorphological properties of geohazard. Results of this study highlight the importance of LiDAR data for creating complete and accurate geohazard inventories, which can then be used for the production of reliable susceptibility and hazard maps and thus contribute to a better understanding of the movement processes and reducing related losses.

Rainy weather greatly affects the visibility of salient objects and scenes in the captured images and videos. The object/scene visibility varies with the type of raindrops, i.e. adherent rain droplets, streaks, rain, mist, etc. Moreover, they pose multifaceted challenges to detect and remove the raindrops to reconstruct the rain-free image for higher-level tasks like object detection, road segmentation etc. Recently, both Convolutional Neural Networks (CNN) and Generative Adversarial Network (GAN) based models have been designed to remove rain droplets from a single image by dealing with it as an image to image mapping problem. However, most of them fail to capture the complexities of the task, create blurry output, or are not time efficient. GANs are a prime candidate for solving this problem as they are extremely effective in learning image maps without harsh overfitting. In this paper, we design a simple yet effective ‘DerainGAN’ framework to achieve improved deraining performance over the existing state-of-the-art methods. The learning is based on a Wasserstein GAN and perceptual loss incorporated into the architecture. We empirically analyze the effect of different parameter choices to train the model for better optimization. We also identify the strengths and limitations of various components for single image deraining by performing multiple ablation studies on our model. The robustness of the proposed method is evaluated over two synthetic and one real-world rainy image datasets using Peak Signal to Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) values. The proposed DerainGAN significantly outperforms almost all state-of-the-art models in Rain100L and Rain700 datasets, both in semantic and visual appearance, achieving SSIM of 0.8201 and PSNR 24.15 in Rain700 and SSIM of 0.8701 and PSNR of 28.30 in Rain100L. This accounts for an average improvement of 10 percent in PSNR and 20 percent in SSIM over benchmarked methods. Moreover, the DerainGAN is one of the fastest methods in terms of time taken to process the image, giving it over 0.1 to 150 seconds of advantage in some cases.

Image authentication based on image hashing has gained large attention in recent years. However, limited work has been done in color image hashing. Also, most of the existing methods are unable to detect tampering, if the composite rotation–scaling–translation (RST) distortion and tampering in a color image occur simultaneously. In this paper, an image hashing technique has been proposed based on convolutional stacked denoising auto-encoders (CSDAEs). In addition, a blind geometric correction approach is used to correct the composite RST distortion in the image. An input image is hierarchically mapped to a lower-dimensional hash code via CSDAEs, which have been trained for content-preserving operations (CPOs). An image map is generated from the hash via the decoder. The tampered area has been localized, by comparing the image map of hash codes from the reference image and the received image. The experimental results show that the proposed method is robust against most of the CPOs, especially to composite RST, a better trade-off between robustness and discrimination, and can localize the tampered regions. The receiver operating characteristics show that the proposed model is better than some of the state-of-the-art methods.

The use of magnetic resonance imaging (MRI) for radiotherapy treatment planning (RTP) is increasing due to excellent tissue contrast and high-resolution images compared to other modalities. However, the issue of image inhomogeneity which affects tissue delineation and contrast during treatment planning needs to be resolved. This study is aimed to qualitatively observe image inhomogeneity in T2-weighted images and T2 measurement using head phantom. Agar hydrogel phantom of various concentrations were used to form inhomogeneous phantom for mimicking tissue variation in the brain in addition to American College of Radiology accredited phantom. T2 images from both phantoms were obtained using fast spin echo sequence with variation of echo time (TE) and repetition time. Image processing was processed to obtain signal intensity and image mapping and the effect of inhomogeneity was observed qualitatively. T2 values were obtained by fitting signal intensity against TE and T2-mapping images were obtained. T2-weighted and T2-mapping images from the experiment provided qualitative description on image inhomogeneity and contrast. The result showed that T2-mapping images offered more information on phantom properties and described image inhomogeneity and contrast better than T2-weighted images. T2 measurement provided an alternative method for better description of image contrast which could enhance localization and tissue delineation in MRI-based RTP.

Blind people can benefit greatly from a system capable of localising and reading comprehension text embedded in natural scenes and providing useful information that boosts their self-esteem and autonomy in everyday situations. Regardless of the fact that existing optical character recognition programmes seem to be quick and effective, the majority of them are not able to correctly recognise text embedded in usual panorama images. The methodology described in this paper is to localise textual image regions and pre-process them using the naïve Bayesian algorithm. A weighted reading technique is used to generate the correct text data from the complicated image regions. Usually, images hold some disturbance as a result of the fact that filtration is proposed during the early pre-processing step. To restore the image's quality, the input image is processed employing gradient and contrast image methods. Following that, the contrast of the source images would be enhanced using an adaptive image map. The stroke width transform, Gabor’s transform, and weighted naïve Bayesian classifier methodologies have been used in complicated degraded images to segment, feature extraction, and detect textual and non-textual elements. Finally, to identify categorised textual data, the confluence of deep neural networks and particle swarm optimisation is being used. The text in the image is transformed into an acoustic output after identification. The dataset IIIT5K is used for the development portion, and the performance of the suggested come up is evaluated using parameters such as accuracy, recall, precision, and F1-score.

Minimizing costs and increasing sales are a goal for every busi- ness nowadays. This fact, together with the development of new technologies, have driven the emergence of virtual applications where the customers can configure the product they are interested in only interacting with the images where the products appear. Many applications are available on Internet or app stores for this purpose. In all of them, a high realism is required. However, this fact is directly related to a high cost of storage of data and to the difficulty of generating the images of the scenes where the product is exposed. This paper presents a virtual configurator addressed to tile factories that solves these problems maintaining a high realism. The developed application generates the configurable images by rendering 3D modeled environments and the customization is performed taking advantage of the graphics hardware. It is in charge of performing the tiling of any size tiles in real time. The presented image mapping system is based on the real measurements of the walls or floor of the environment that appear in the image and on the dimensions of the tile to map. Taking these data into account, the application performs the final appearance adapting the final image to the requirements of the user. The presented method reduces the amount of stored information maintaining the realism of the customized images.

We introduce a novelty in the method of the deep salient wood image quality assessment (DS-WIQA) for no-reference image quality assessment (NR-IQA). We exploit a five-layer deep convolutional neural network (DCNN) for the salient wood image map. DS-WIQA also employs the n-convex-concave model. The outcomes obviously prove that our DCNN and DS-WIQA architectures can deliver a superior achievement on Zenodo and Lignoindo datasets, respectively. We compute a salient wood image map of each wood image in small wood image patches. Our exploratory outcomes evince that the proposed DCNN and DS-WIQA methods are superior to other the advanced methods on Zenodo and Lignoindo datasets, respectively. Our proposed DCNN for NR-IQA also obtains a better result compared with the other NR-IQA methods in the five distortion types of JP2K, JPEG, white noise Gaussian, blocking artifact, and the fast fading and also in the undistorted wood images. Our DCNN outruns the recent most sophisticated methods in terms of SROCC and LCC evaluation, respectively. DS-WIQA outpaces other the advanced methods by $$0.38\%$$ and $$0.22\%$$ greater than our proposed DCNN, and $$34.84\%$$ and $$30.15\%$$ greater than other methods with respect to SROCC and LCC, respectively. In computational complexity of our proposed DCNN and DS-WIQA cut down the shift add operation in exponential, logarithmic, and trigonometric functions. DS-WIQA shows up to be more significant than our proposed DCNN and the other DCNN methods.

In this paper, we present an image classification approach to action recognition with 3D skeleton videos. First, we propose a video domain translation-scale invariant image mapping, which transforms the 3D skeleton videos to color images, namely skeleton images. Second, a multi-scale dilated convolutional neural network (CNN) is designed for the classification of the skeleton images. Our multi-scale dilated CNN model could effectively improve the frequency adaptiveness and exploit the discriminative temporal-spatial cues for the skeleton images. Even though the skeleton images are very different from natural images, we show that the fine-tuning strategy still works well. Furthermore, we propose different kinds of data augmentation strategies to improve the generalization and robustness of our method. Experimental results on popular benchmark datasets such as NTU RGB + D, UTD-MHAD, MSRC-12 and G3D demonstrate the superiority of our approach, which outperforms the state-of-the-art methods by a large margin.

In this paper, a novel framework, named as global-local feature attention network with reranking strategy (GLAN-RS), is presented for image captioning task. Rather than only adopting unitary visual information in the classical models, GLAN-RS explores the attention mechanism to capture local convolutional salient image maps. Furthermore, we adopt reranking strategy to adjust the priority of the candidate captions and select the best one. The proposed model is verified using the Microsoft Common Objects in Context (MSCOCO) benchmark dataset across seven standard evaluation metrics. Experimental results show that GLAN-RS significantly outperforms the state-of-the-art approaches, such as multimodal recurrent neural network (MRNN) and Google NIC, which gets an improvement of 20% in terms of BLEU4 score and 13 points in terms of CIDER score.

This paper reviews more than 10 recent doctoral theses and master’s theses accomplished at the Chair of Cartography, Technical University of Munich (TUM). These research works address two persistent challenges involved in the digital mapmaking process: understanding complex input data and improving cartographic design to keep up with the changing roles and increasing demands of users on immediately usable maps. At first, a dedicated computational approach based on Conditional Random Fields (CRF) for data enhancement of Volunteered Geographic Information (VGI) is introduced and demonstrated with test data from trajectories of floating taxis and OpenStreetMap (OSM). It is followed by a summarized overview of visual analytical approaches for event and behavior discovery and their implementations on various VGI data sources including trajectories of floating taxis, twitter messages and trajectories of football matches. Finally, a concept of image map with a multilayered visual hierarchy is proposed and demonstrated with a set of attention-guided design strategies. Keywords: Probabilistic Graphical Model, visual analytics, event and behavior discovery, attention-guided design Der vorliegende Beitrag gibt einen Uberblick uber die Forschungsinhalte von mehr als 10 aktuellen Dissertationen und Masterarbeiten, die am Lehrstuhl fur Kartographie der Technischen Universitat Munchen durchgefuhrt wurden. Diese Arbeiten befassen sich mit zwei dauerhaften Herausforderun- gen in der digitalen Kartographie: Verstehen von komplexen Eingabedaten und Verbesserung der Gestaltungsverfahren, um mit den sich verandernden Rollen und den stetig wachsenden Anforderungen der Kartennutzer mitzuhalten. Zunachst wird ein auf Probabilistischen Graphischen Modellen (PGM) basierter Ansatz zur Datenanreicherung der VGI vorgestellt und anhand Taxifahrtrouten und Strasennetz von OpenStreetMap (OSM) demonstriert. Es folgt eine Darstellung von visuellen analy- tischen Ansatzen sowie deren Umsetzung zur Entdeckung des Events und des Verhaltens von Personen in verschiedenen VGI-Datenquellen wie z.B. Trajektorien der Floating Taxis, Twitternachrichten und Trajektorien des Fusballspiels. Schlieslich wird das Konzept der Bildkarte, die in der Regel aus Kartensymbolen im Vordergrund und einem Rasterbild im Hintergrund besteht, in Frage gestellt. Eine Reihe von Aufmerk- samkeitsgetriebenen Gestaltungsstrategien zeigen, wie sich Bildkarten mit mehreren visuellen Ebenen generieren lassen. Schlusselworter: Probabilistische Graphische Modelle, Visuelle Analytik, Entdeckung des Events und Verhaltens, Aufmerksamkeits-betriebene Gestaltung

Image registration (IR) aims to map one image to another of a same scene. With rapid progress in image acquisition technologies, 3D IR becomes an important problem in magnetic resonance imaging (MRI) and other applications. In the literature, however, most IR methods are for 2D images and there are only a limited number of 3D methods available. Because 3D images have much complicated structure than their 2D counterparts, 3D IR is not just a simple generalization of the 2D IR problem. In this paper, we develop a 3D IR method that can handle cases with affine geometric transformations well. By its definition, an affine transformation maps a line to a line, and it includes rotation, translation, and scaling as special cases. In practice, most geometric transformations involved in IR problems are affine transformations. Therefore, our proposed method can find many IR applications. It is shown that this method works well in various cases, including cases when the data size of a 3D image is reduced for different reasons. This latter property makes it attractive for many 3D IR applications, since 3D images are often big in data size and it is natural to reduce their size for fast computation.

Small tunnels such as engineering geological exploratory tunnels and mine roadways are generally narrow, which make the existing photographic geological logging technique inapplicable. Therefore, geological logging of exploratory tunnels has always been taking the method of manual sketch work which has low efficiency and poor informatization degree of products, and it is a technical issue requiring urgent settlement for geological logging of small tunnels. This paper proposes and studies novel methods of photographic geological logging suitable for small tunnels, including image acquisition, image orientation control, image geometric correction, unfolded image map generation and geological attitude measurement, etc. Experiments show that the method can meet the precision requirement of geological logging. The novel method helps to realize the fast acquisition and processing of image-based geological logging data for small tunnels, and the forms of logging result are more abundant and more applicable to informatized management and application of geological logging data.

Surgical navigation technology directed at fetoscopic procedures is relatively underdeveloped compared with other forms of endoscopy. The narrow fetoscopic field of views and the vast vascular network on the placenta make examination and photocoagulation treatment of twin-to-twin transfusion syndrome challenging. Though ultrasonography is used for intraoperative guidance, its navigational ability is not fully exploited. This work aims to integrate 3D ultrasound imaging and endoscopic vision seamlessly for placental vasculature mapping through a self-contained framework without external navigational devices. This is achieved through development, integration, and experimentation of novel navigational modules. Firstly, a framework design that addresses the current limitations based on identified gaps is conceptualized. Secondly, integration of navigational modules including (1) ultrasound-based localization, (2) image alignment, and (3) vision-based tracking to update the scene texture map is implemented. This updated texture map is projected to an ultrasound-constructed 3D model for photorealistic texturing of the 3D scene creating a panoramic view of the moving fetoscope. In addition, a collaborative scheme for the integration of the modular workflow system is proposed to schedule updates in a systematic fashion. Finally, experiments are carried out to evaluate each modular variation and an integrated collaborative scheme of the framework. The modules and the collaborative scheme are evaluated through a series of phantom experiments with controlled trajectories for repeatability. The collaborative framework demonstrated the best accuracy (5.2 % RMS error) compared with all the three single-module variations during the experiment. Validation on an ex vivo monkey placenta shows visual continuity of the freehand fetoscopic panorama. The proposed developed collaborative framework and the evaluation study of the framework variations provide analytical insights for effective integration of ultrasonography and endoscopy. This contributes to the development of navigation techniques in fetoscopic procedures and can potentially be extended to other applications in intraoperative imaging.

The present study utilizes the processed SPOT 5 data to discriminate and to generate 1:10,000 geologic image map to delineate the mineralized diorite–tonalite intrusion around Bulghah gold mine area, Saudi Arabia. The rock units exposed in the area include gossan, marble, Hulayfah volcanics, diorite–tonalite, gneissose granite, and alkali granite. Gold mineralization at Bulghah mine is hosted mainly by Syn- to Late-tectonic diorite–tonalite intrusion aligned along N–S direction and is associated mainly with cataclastic zones and quartz veins. Gossans and jasparoidal gossans (Hulayfah group), recorded at the western side of Bulghah mine area as discontinuous small lenses, can be easily discriminated on 4/2 and 4/3 band ratio SPOT images by their white and black image signatures, respectively. Granitoids (gneissose granite and alkali granite) are easily discriminated in 3/2 ratio image, in which gneissose granite has gray image signature, whereas alkali granite has dark gray image signature. On the SPOT false color composite band ratios image (3/2 R, 4/2 G, and 4/3 B), gossan, marble, Hulayfah volcanics, diorite–tonalite, gneissose granite, and alkali granite have sky blue, blood red, bluish light brown, orange, brick red, and deep blue colors, respectively. Fusion of the false color composite SPOT ratios image (3/2 R, 4/2 G, and 4/3 B) with the high spatial resolution SPOT pan image is performed using IHS transformation method. The fused image is used to delineate the mineralized diorite–tonalite intrusion and to produce 1:10,000 geologic image map for Bulghah gold mine area. The present study reveals the usefulness of the processed SPOT 5 data for adding new extensions at the southern and northern boundaries of diorite–tonalite intrusion.

Image maps have become very popular and frequently produced cartographic outputs during recent years. However, the unambiguous terminology, definitions, content and appearance specification have not been widely researched. The paper deals with the new definition of image map, its component delineation, and basic classification. The authors present aspects of topographic and thematic image maps. The main theoretical achievement of the authors’ research is a formulation of the image component and the symbol component of image map content. Finally, several examples of topographic and thematic image maps are presented. They differ according to the relationship between topographic background and thematic content, and between image and symbol component. The image component can be a carrier of thematic information, or it can be used as topographic background. The image map examples were compiled for use by the military, urban planners and natural risk emergency services etc. All examples demonstrate how to design, complete and use image-based cartographic products. Those variants might be used as guidelines for future image map production, especially for governmental and research purposes.

This paper describes a task-based dynamic geometry platform that is able to record student responses in a collective fashion to pre-designed dragging tasks. The platform provides a new type of data and opens up a quantitative dimension to interpret students’ geometrical perception in dynamic geometry environments. The platform is capable of generating a collective image map of student geometrical perceptions for a pre-designed dragging task. This map is interpreted as students’ qualitatively different ways of perceiving a geometrical phenomenon under the drag mode, ways which are quantified and categorized in a collective way. The idea of task perceptual landscape is proposed to facilitate discussion on the pedagogical significance of this platform. Specifically, a task case is presented and analysed in which a methodology is developed that provides a way to classify students’ geometrical perceptions with respect to the task. The task perceptual landscape is interpreted as a collective example space of student perception of a task. Furthermore, an idea of personal example space is developed through the findings from a qualitative study for the same task. This brings about discussion on possible pedagogical correlation between the quantitative and qualitative aspects of the platform.

A novel automatic seamless stitching method is presented. Compared to the traditional method, it can speed the processing and minimize the utilization of human resources to produce global lunar map. Meanwhile, a new global image map of the Moon with spatial resolution of ∼120 m has been completed by the proposed method from Chang’E-1 CCD image data.

Existing content-based image retrieval paradigms almost never address the problem of starting the search, when the user has no starting example image but rather a mental image. We propose a new image retrieval system to allow the user to perform mental image search by formulating boolean composition of region categories. The query interface is a region photometric thesaurus which can be viewed as a visual summary of salient regions available in the database. It is generated from the unsupervised clustering of regions with similar visual content into categories. In this thesaurus, the user simply selects the types of regions which should and should not be present in the mental image (boolean composition). The natural use of inverted tables on the region category labels enables powerful boolean search and very fast retrieval in large image databases. The process of query and search of images relates to that of documents with Google. The indexing scheme is fully unsupervised and the query mode requires minimal user interaction (no example image to provide, no sketch to draw). We demonstrate the feasibility of such a framework to reach the user mental target image with two applications: a photo-agency scenario on Corel Photostock and a TV news scenario. Perspectives will be proposed for this simple and innovative framework, which should motivate further development in various research areas.

Multimedia digital libraries are gaining an increasing relevance, and are growing in both number and size. Moreover, advancements in acquisition and authoring technologies have also caused new media types to be added to these libraries. For instance, animations and 3D models have recently come alongside more traditional media such as text, images, audio and video. In particular, even if 3D objects were initially included in multimedia digital libraries in the form of images (i.e., views of the objects), they are now also available in the form of 3D models. These allow to overcome the inherent limitations of an image-based representation of a three-dimensional reality. In this paper we present a novel method for description and retrieval of 3D models that relies on curvature maps: after an initial pre-processing of the model, differential properties of points on the surface of the 3D object are evaluated; the model surface is then deformed into an ellipsoid, and is mapped onto a 2D image retaining curvature information of the original model. Through this media conversion from 3D model to image map, we enable application of analysis, description, and matching techniques which have been originally conceived for the image medium. In particular, the two approaches described in this paper rely on histograms for a coarse but efficient description of tiles of the curvature map, and on weighted walkthroughs for a more precise description of homogeneous regions of the map, taking into account their spatial arrangement and areas. Experimental results reveal that both techniques can be successfully applied to curvature maps. The two approaches have been compared against other solutions presented in literature, and have shown a good retrieval performance (particularly in the case of region-based description).

Because digital images are not meaningful by themselves, images are often coupled with some descriptive or qualitative data in an image database. These data, divided into syntactic (color, shape, and texture) and semantic (meaningful real word object or concept) features, necessitate novel querying techniques. Most image systems and prototypes have focussed on similarity searches based upon the syntactic features. In the DISIMA system, we proposed an object-oriented image data model that introduces two main types: image (that represents an image and its descriptive properties) and salient object (that represents the semantics of an image). We further defined operations on the images and the salient objects as new joins. This approach is necessary in order to envision a declarative query language for images. This paper summarizes the querying facilities implemented for the DISIMA system and gives their theoretical foundation: the data model and the complementary algebraic operations, the textual query language (MOQL) and its visual counterpart (VisualMOQL) based on an image calculus. Both languages are declarative and allow the combination of semantic and similarity queries.

Digital as well as photographic satellite image data offer a high potential of topographic and thematic information. The paper discusses the application of such data for mapping purposes. For the compilation of conventionalTopographic Maps the requirements concerning the geometrical accuracy can easily be met for 1 : 50,000 or even 1 : 25,000. However, the interpretability of the features that are relevant for topography is often the critical factor. Only the Russian KWR-1000 data are suitable for mapping topographic features in 1 : 25,000. Main emphasis is laid on a new map type, theSatellite Image Map. The authors describe the satellite image mapping procedures developed at the Technical University of Berlin and applied in practice by FPK Consulting Engineers. Some related problems are discussed in details, eg mosaicking, merging of different types of data, and integration of graphical elements. An increase in cartographic applications of satellite remote sensing data is anticipated.

The practice has proved that it is an economic and effective method to investigate placer gold deposit by using multi-level information sources of remote sensing and multi-variate analysis methods, especially for the area with a sparse population and difficult condition like the Da Hinggan Mountains, China. The information sources used in our work includes Landsat TM, aerial infrared photography and their mosaic image maps and enlarged photos with different scales. According to statistic data, in the study area the gold-bearing rocks are mainly granite, alaskite, granodiorite and some old metamorphic rocks. On gold-bearing geological structures, the fault zones in the four directions (NE, NNE, NW and EW) are obvious, in which NNE and EW are the most key fault zones. On fluvial geomorphology the flow courses stored placer are in the tributaries of the 4th and 5th levels, especially in straight or slight curve reaches. On the basis of analysis the interpretative signs were set up, and the interpretative work began on a sheet of picture with a single band at first, furtherly, the integral interpretation on color enlargement pictures the extraction and composition work was carried out by means of multicolor digital analyzer, and computer image processing equipment with the VAX 11 / 750 as a host computer so as to draw up the relatively specialized maps of placer deposit and delimit the locations of ore body on the pictures or on the screen. Finally, the satisfatory effect was achieved through a series of comprehensive analyses.

Color-image maps are generated through computer-based simulations to study the evolution of a system of three bodies interacting under classical gravitational forces. These image maps partition the set of possible initial-system configurations into regions that exhibit similar evolutionary characteristics. This resembles the technique used to produce color images of the Mandelbrot set. A restricted case of the three-body problem, in which the initial velocities of all three bodies are zero and the bodies are all of equal mass, is examined in detail. Image maps of the time to collision (assuming a finite radius) as a function of initial configuration are presented. Symmetries of the resulting maps are analyzed for invariants by means of transformations of scale.