Abstract: Detecting tables in document images is important since not only do tables contain important information, but also most of the layout analysis methods fail in the presence of tables in the document image. Existing approaches for table detection mainly focus on detecting tables in single columns of text and do not work reliably on documents with varying layouts. This paper presents a practical algorithm for table detection that works with a high accuracy on documents with varying layouts (company reports, newspaper articles, magazine pages, ...). An open source implementation of the algorithm is provided as part of the Tesseract OCR engine. Evaluation of the algorithm on document images from publicly available UNLV dataset shows competitive performance in comparison to the table detection module of a commercial OCR system.

Abstract: In this article, a system to detect rooms in architectural floor plan images is described. We first present a primitive extraction algorithm for line detection. It is based on an original coupling of classical Hough transform with image vectorization in order to perform robust and efficient line detection. We show how the lines that satisfy some graphical arrangements are combined into walls. We also present the way we detect some door hypothesis thanks to the extraction of arcs. Walls and door hypothesis are then used by our room segmentation strategy; it consists in recursively decomposing the image until getting nearly convex regions. The notion of convexity is difficult to quantify, and the selection of separation lines between regions can also be rough. We take advantage of knowledge associated to architectural floor plans in order to obtain mostly rectangular rooms. Qualitative and quantitative evaluations performed on a corpus of real documents show promising results.

Abstract: Table spotting and structural analysis are just a small fraction of tasks relevant when speaking of table analysis. Today, quite a large number of different approaches facing these tasks have been described in literature or are available as part of commercial OCR systems that claim to deal with tables on the scanned documents and to treat them accordingly.However, the problem of detecting tables is not yet solved at all. Different approaches have different strengths and weak points. Some fail in certain situations or layouts where others perform better. How shall one know, which approach or system is the best for his specific job? The answer to this question raises the demand for an objective comparison of different approaches which address the same task of spotting tables and recognizing their structure.This paper describes our approach towards establishing a complete and publicly available, hence open environment for the benchmarking of table spotting and structural analysis. We provide free access to the ground truthing tool and evaluation mechanism described in this paper, describe the ideas behind and we also provide ground truth for the 547 documents of the UNLV and UW-3 datasets that contain tables.In addition, we applied the quality measures to the results that were generated by the T-Recs system which we developed some years ago and which we started to further advance since a few months.

Abstract: Handwriting recognition in historical documents is vital for the creation of digital libraries. The creation of readily available ground truth data plays a central role for the development of new recognition technologies. For historical documents, ground truth creation is more difficult and time-consuming when compared with modern documents. In this paper, we present a semi-automatic ground truth creation proceeding for historical documents that takes into account noisy background and transcription alignment. The proposed ground truth creation is demonstrated for the IAM Historical Handwriting Database (IAM-HistDB) that is currently under construction and will include several hundred Old German manuscripts. With a small set of algorithmic tools and few manual interactions, it is shown how laypersons can efficiently create a ground truth for handwriting recognition.

Abstract: Segmentation of a document image into text and non-text regions is an important preprocessing step for a variety of document image analysis tasks, like improving OCR, document compression etc. Most of the state-of-the-art document image segmentation approaches perform segmentation using pixel-based or zone(block)-based classification. Pixel-based classification approaches are time consuming, whereas block-based methods heavily depend on the accuracy of block segmentation step. In contrast to the state-of-the-art document image segmentation approaches, our segmentation approach introduces connected component based classification, thereby not requiring a block segmentation beforehand. Here we train a self-tunable multi-layer perceptron (MLP) classifier for distinguishing between text and non-text connected components using shape and context information as a feature vector. Experimental results prove the effectiveness of our proposed algorithm. We have evaluated our method on subset of UW-III, ICDAR 2009 page segmentation competition test images and circuit diagrams datasets and compared its results with the state-of-the-art leptonica's page segmentation algorithm.

Abstract: This paper provides an analysis of the BADA aircraft performance model capabilities and addresses the BADA model ability to provide accurate modeling of aircraft performances over the complete flight envelope for a number of aircraft types and different ways in which an aircraft can be operated during the flight. The focus of the paper is the support of complex aircraft operations by BADA. A short description of the two existing BADA families and their main characteristics is given. The complex flight instructions and operating regimes — economy climb, cruise and descent based on cost index, maximum range cruise, long range cruise, optimum altitude and maximum endurance cruise — identified as key features in support to optimized flight execution are discussed. The optimization procedures and equations in which they derive are presented and the ability of the BADA model to support these flight operations is demonstrated. It is shown that BADA 4 can be successfully used with complex instructions and operating regimes, whereas the use of BADA 3 is limited. Finally, the results of a validation experiment dedicated to BADA thrust models are presented.

Abstract: Many feature extraction approaches for off-line handwriting recognition (OHR) rely on accurate binarization of gray-level images. However, high-quality binarization of most real-world documents is extremely difficult due to varying characteristics of noises artifacts common in such documents. Unlike most of these features, Gabor features do not require binarization of the document images, and thus are likely to be more robust to noises in document images. To demonstrate the efficacy of our proposed Gabor features, we perform subword recognition for off-line Arabic handwritten images using Support Vector Machines (SVM). We also compare the recognition performance with other binarization based features which have been proven to be effective in capturing shape characteristics of handwritten Arabic subwords, such as GSC (a set of gradient, structure, and concavity features) and skeleton based Graph features. Our preliminary experimental results show that Gabor features outperform Graph features and are slightly better than GSC features for Arabic subword recognition. In addition, by combining Gabor and GSC features, we obtain a significant reduction in classification error rate over using GSC or Gabor features alone.

Abstract: We have developed an early prototype of an ARINC 653 implementation using the virtualization technology of the open source Xen hypervisor along with a Linux-based domain/partition OS. In this paper we share lessons learned from adding to our prototype both an ARINC 653 CPU scheduler and a simple ARINC 653 serial I/O driver. By using a common hypervisor technology on multiple platforms, early application development can be done in a PC environment with relatively good modeling of the final target's behavior. The paper covers three topics. First, we start with an overview of the ARINC 653 standard, which is important because it reduces development costs, reduces system weight, and lowers certification costs. The standard focuses on resource partitioning of time and space on an avionics computer, managing the three primary subsystems of the computer: Central Processing Unit (CPU), Memory, and Input/Output (I/O). Second, we will review virtualization technology, an established method of sharing a computing resource, considering it for adaptation to ARINC 653. Third, as a case study, we will examine our prototype implementation of the ARINC 653 standard using the Xen open source hypervisor. We conclude with a discussion of our plans for future work towards ARINC 653 simulation and development environments on both PC desktop and embedded targets.

Abstract: A novel text extraction method from graphical document images is presented in this paper. Graphical document images containing text and graphics components are considered as two-dimensional signals by which text and graphics have different morphological characteristics. The proposed algorithm relies upon a sparse representation framework with two appropriately chosen discriminative overcomplete dictionaries, each one gives sparse representation over one type of signal and non-sparse representation over the other. Separation of text and graphics components is obtained by promoting sparse representation of input images in these two dictionaries. Some heuristic rules are used for grouping text components into text strings in post-processing steps. The proposed method overcomes the problem of touching between text and graphics. Preliminary experiments show some promising results on different types of document.

Abstract: Continuous Descent Approaches (CDAs) can significantly reduce fuel burn and noise impact by keeping arriving aircraft at their cruise altitude for longer than during conventional approaches(to descend as late as possible)and then having them make a continuous descent to the runway at near idle thrust with no level flight segments. The CDA procedures are fixed routes that are vertically optimized. With the changing traffic conditions and variable noise abatement rules the benefits of CDA operations are not yet fully realized. In this paper we propose a methodology to generate aircraft-specific dynamic CDA routes that are both laterally and vertically optimized on given objectives (noise, emission and fuel) from an Initial Approach Fix (IAF) to Final Approach Fix (FAF). The methodology utilizes real-time aircraft position and defined objectives to generate CDA routes which can then be converted into a set of artificial waypoints for continuous descent in transition airspace. The methodology involves discretizing the terminal airspace into concentric cylinders with artificial waypoints and uses enumeration and elimination (based on aircraft performance envelope) from one waypoint to other to identify all the possible routes. For each transition a variety of metrics including noise, emission and fuel burn are computed. From the resulting set of possible CDA routes, those routes are identified that represent the best trade-off on the given objectives. One of these routes is then used to dynamically update the flight route for executing the CDA procedure. For noise we used The Overall Sound Pressure Level (OPSL) and for emissions we used four pollutants HC, CO, CO 2 and NO x . The dynamic CDA algorithm is implemented in a high-fidelity simulator ATOMS for Sydney Terminal Area with 34L as arrival runway for a Melbourne-Sydney flight (B737–400 aircraft, CFM56–3C–1 engines with a nominal weight of 58000 kg). The dynamic CDA routes are then compared on noise, emission and fuel burn with same flight conducting a typical CDA procedure (MANFA ONE Arrival) at the Sydney airport. The results shows that the methodology generates 64 possible solutions (dynamic CDA routes) from IAF to FAF in the transition airspace, of which 5 solutions were non-dominated. Dynamic CDA approach shows a reduction of 14.96% in noise, 11.6% reduction in NO x emission and 1.5% reduction in fuel burn when compared to a standard CDA trajectory. The paper also investigates the throughput capacity of transition airspace for multiple flights performing CDA operation. The methodology incorporates a delay algorithm which uses the flights' estimated time of arrival (ETA) at the IAF and then allocates them a conflict free CDA route by searching through available routes. The approach takes into account the aircraft category and corresponding time occupancy at each artificial waypoint of the proposed CDA routes and propagate delays back when conflict exists.

Abstract: On-line handwriting recognition systems are usually better than their off-line counterparts thanks to the accessibility of dynamic information such as stroke order, velocity, acceleration, and pressure. Whilst the exact value of velocity as well as acceleration or pressure is unlikely to be recoverable, the temporal order of the strokes or the pen trajectory is shown to be more promising for recovery. The published experimental results suggest that the recovered pen trajectory information actually improves the off-line recognition accuracy. This paper presents an overview and discussion of pen trajectory recovery methods developed to date.

Abstract: Modeling aircraft taxi operations is an important element in understanding current airport performance and where opportunities may lie for improvements. A statistical learning approach to modeling aircraft taxi time is presented in this paper. This approach allows efficient identification of relatively simple and easily interpretable models of aircraft taxi time, which are shown to yield remarkably accurate predictions when tested on actual data.

Abstract: In this paper we present a method for organizing and indexing logo digital libraries like the ones of the patent and trademark offices. We propose an efficient queried-by-example retrieval system which is able to retrieve logos by similarity from large databases of logo images. Logos are compactly described by a variant of the shape context descriptor. These descriptors are then indexed by a locality-sensitive hashing data structure aiming to perform approximate k-NN search in high dimensional spaces in sub-linear time. The experiments demonstrate the effectiveness and efficiency of this system on realistic datasets as the Tobacco-800 logo database.

Abstract: This paper describes the steps that have been undertaken in order to develop the IBN SINA database, which is designed to apply learning techniques in the processing and understanding of document images. The description of the preparation process, including preprocessing, feature extraction and labeling, is provided. The database has been evaluated using classification techniques, such as the SVM classifiers. In order to make the database compatible with these classifiers, the labels of the shapes have been translated into a set of bi-class problems. Promising results with the SVM classifiers have been obtained.

Abstract: In modern aircrafts, more and more functions traditionally implemented as Line Replaceable Units (LRUs) will be hosted by Integrated Modular Avionics (IMA) modules. At the same time new aircraft programs will require new safety functions, information services and comfort features which will also increase the demand for processing performance of IMA modules. The traditional approach to provide more processing bandwidth was to increase the CPU clock frequency, increase pseudo-parallel processing on instruction level through instruction pipelines and speculative program executions and to increase the cache size and number of cache levels. With today's technology, this approach has reached its limit. Increasing CPU frequency causes disproportionate power consumption and thermal dissipation loss and raises more and more problems due to chip internal and external crosstalk, signal delays and reflection. Existing parallelism and dependencies on code level prevent further performance improvement through parallel execution on instruction level. To further increase processor performance, the chip industry has switched to a multi-core design for the high performance processors. The development of multi-core based high performance IMA platforms will be a necessary step to reach a larger scale integration on function level. The question is, "Can a multi-core based platform reach the same level of determinism as a single core platform and can this be demonstrated?" This paper addresses certification aspects of multi-core based IMA platforms with the focus on today's technologies and processes. The paper provides an analysis of potential hardware and software related interference channels between partitions running on a multi-core based platform. Different core software concepts found in existing implementations like asymmetric multi processing (AMP) and symmetric multi processing (SMP) concepts are evaluated with respect to partitioning aspects.

Abstract: In this paper, we present a new predictive model for estimating airport delay using data from weather forecast products We use the well established Weather Impacted Traffic Index (WITI) toolset and metric The latter quantifies the “front-end” impact of weather and traffic demand on the NAS and is well correlated with NAS delays, which makes WITI a reasonably good high-level model of NAS performance WITI-FA (“Forecast Accuracy”) is the forecast-weather counterpart to WITI: it can use various convective forecast products, as well as Terminal Area Forecast (TAF), and quantify forecast weather impact on the NAS We show how these models can be refined and re-oriented toward predictive capability First, instead of using just three WITI components, we break down the weather impacts by type, eg wind, snow, low ceilings, en-route thunderstorms, volume, etc — twelve components in total Second, instead of using a NAS-wide WITI model, we “zoom in” on individual airports The model is calibrated to minutes of delay for a given airport on an hourly basis Having trained the model using historical airport performance and actual weather / scheduled traffic data, we then apply it in a predictive mode The paper contains multiple examples and comparisons of predicted vs actual delays at major airports under various weather conditions In addition to predicting delays, the model can be used as a decision support tool If predicted delays are too high, WITI can be run in what-if mode to gauge demand reduction, guaranteeing sustainable delays in adverse weather conditions This could also be helpful to airlines when they need to decide on the amount of flight cancellations Lastly, our airport delay predictor model can be used to compare the efficacy of different weather forecast products

Abstract: Trajectory Based Operations (TBO) is a key component of both the US Next Generation Air Transport System (NextGen) and Europe's Single European Sky ATM Research (SESAR). There is a significant amount of effort underway in both programs to advance this concept. Trajectory Synchronization and Negotiation are key required capabilities in both the NextGen and SESAR TBO concepts, and they provide the framework to improve the efficiency of airspace operations. In recognition of the importance of TBO, General Electric and Lockheed Martin have created a Joint Strategic Research Initiative (JSRI), which aims to generate technologies that accelerate adoption of TBO. This paper explores various trajectory synchronization and negotiation concepts, including existing gaps and shortfalls. This paper also presents the JSRI simulation and evaluation environment being developed, which embeds trajectory synchronization and negotiation concepts, and has the potential to address existing gaps and shortfalls.

Abstract: Conventional optical character recognition (OCR) systems operate on individual characters and words, and do not normally exploit document or collection context. We describe a Collection OCR which takes advantage of the fact that multiple examples of the same word (often in the same font) may occur in a document or collection. The idea here is that an OCR or a reCAPTCHA like process generates a partial set of recognized words. In the second stage, a nearest neighbor algorithm compares the remaining word-images to those already recognized and propagates labels from the nearest neighbors. It is shown that by using an approximate fast nearest neighbor algorithm based on Hierarchical K-Means (HKM), we can do this accurately and efficiently. It is also shown that profile based features perform much better than SIFT and Pyramid Histogram of Gradient (PHOG) features. We believe that this is because profile features are more robust to word degradations (common in our documents). This approach is applied to a collection of Telugu books - a language for which no commercial OCR exists. We show from a selection of 33 Telugu books that starting with OCR labels for only 30% of the collection we can recognize the remaining 70% of the words in the collection with 70% accuracy using this approach. Since the approach makes no language specific assumptions, it should be applicable to a large number of languages. In particular we are interested in its applicability to Indic languages and scripts.

Abstract: This paper proposes a new MAC protocol applied for a new kind of data collection applications that use a wireless sensor network employed with one unmanned aerial vehicle (WSN-UAV). In this protocol, the sensors are classified in different groups based on priorities and communicate with the UAV by a CDMA-based transmission scheme. The Prioritized Frame Selection based CDMA MAC protocol (PFSC-MAC1) incorporates not only the value, but also the variation trend of the receiving power of the beacon signals. This protocol provides a low rate of failed packet due to the mobility of the UAV, which is the most critical metric in these types of applications. In addition, an optimal number of priority groups have also been introduced to guarantee a minimal transmission interval and a lowest failed packet ratio. Numerical results have also shown the efficient energy compared with the other protocol that is based on TDMA. This protocol is simple but promising for various applications that collect data from a large area sensor network.

Abstract: Energy efficiency, noise and emissions footprint reduction are major “green” performance targets of next-generation air transportation systems. Automatic Dependent Surveillance Broadcast (ADS-B) is an advanced “cyber-physical” concept that can help meet such targets by enabling highly accurate air traffic surveillance. However, beneficial, confident application of ADS-B requires a foundational understanding of vulnerabilities and mitigation requirements. This paper focuses on the assessment and mitigation of threats from unauthorized access and unanticipated disruption of ADS-B communications. We present a solution to protect message integrity in ADS-B IN based airborne surveillance, discuss the accuracy and robustness of the solution in the presence of spoofed or compromised nodes and intentional jamming. To study effects of “intelligent jamming” strategies on ADS-B, visualize and analyze system-wide impact, we propose a security simulation tool concept and design.

Abstract: This paper describes the Collaborative Departure Queue Management (CDQM) concept, which has been developed by Mosaic ATM under funding from the FAA's Surface Trajectory Based Operations (STBO) project and NASA [1–3]. Under CDQM procedures, airport departure capacity is allocated to flight operators and metering procedures are used to manage the length of the runway queue. This management of the runway departure queue allows fuel burn, emissions and airport surface congestion to be reduced. Operational field evaluations of CDQM have been conducted over the last year at the Memphis International Airport. This paper provides initial results of these CDQM field evaluations.

Abstract: Multicore virtualization can offer significant benefits to embedded avionics systems with regard to enabling mixed real-time and guest operating system interoperability, legacy code migration, and hardware consolidation Virtualization enabled architectures have evolved from a traditional Hypervisor Monolithic Model (VmWare and VirtualLogix), to a Hypervisor Console Guest Monolithic Model (Xen), and now to a High Assurance Microkernel Hypervisor RTOS Model The ability to consolidate multiple legacy Single Board Computers (SBCs) with various guest operating systems and applications into a multicore, virtualized SBC is a critical enabler to next generation avionics This paper describes an initial feasibility assessment toward applying the Microkernel Hypervisor RTOS Virtual Machine (VM) architecture to enable virtualization for a representative set of avionics applications requiring multiple guest OS environments The specific notional configuration included: legacy application execution on a legacy RTOS guest OS in VM1, newer application execution on a more recently released level of RTOS on VM2, safety critical applications execution on an ARINC 653 OS on VM3, Global Information Grid (GIG) applications execution on a Linux guest OS on VM4, and MILS/MLS application execution on a high assurance OS on VM5, all executing on a Microkernel Hypervisor RTOS within a Multicore (X86 or Power PC) with hardware-based virtualization support The paper identifies the current system design issues, limitations/restrictions, and feasibility of applying representative products in this representative hybrid legacy/next generation environment The system design challenges identified included: 1) selection of communication mechanisms and scheduling for mixed operating system environments, 2) addressing current limitations/restrictions of current vendor products with regard to multicore, 3) properly scheduling the infrastructure to meet the safety and security requirements, 4) incorporating extensions for MultiLevel Security (MLS) components for networked GIG and local connectivity, and 5) consolidating I/O components without compromising safety, security, and redundancy considerations

Abstract: This paper describes a class of strategies for reducing persistent contrail formation in the United States airspace. The primary objective is to minimize potential contrail formation regions by altering the aircraft's cruising altitude in a fuel-efficient way. The results show that the contrail formations can be reduced significantly without extra fuel consumption and without adversely affecting congestion in the airspace. The contrail formations can be further reduced by using extra fuel. For the day tested, the maximal reduction strategy has a 53% contrail reduction rate. The most fuel-efficient strategy has an 8% reduction rate with 2.86% less fuel-burnt compared to the maximal reduction strategy. Using a cost function which penalizes extra fuel consumed while maximizing the amount of contrail reduction provides a flexible way to trade off between contrail reduction and fuel consumption. It can achieve a 35% contrail reduction rate with only 0.23% extra fuel consumption. The proposed fuel-efficient contrail reduction strategy provides a solution to reduce aviation-induced environmental impact on a daily basis.

Abstract: There is significant potential to decrease fuel burn, emissions, and delays of aircraft at airports by optimizing surface operations. A simple surface traffic optimization approach is to hold aircraft back at the gates based on aggregate information on surface queues. Depending on the level of surface surveillance and onboard equipage, it may also be possible to use a more complex approach, namely, to simultaneously optimize the surface trajectories of all taxiing aircraft. Using data from the Detroit Metropolitan Wayne County airport (DTW), this paper compares the benefits of the two approaches, and finds that at a relatively uncongested airport such as DTW, the aggregate queue-based approach only yields modest improvements in taxi-out time, while the trajectory-based approach yields a nearly 23% decrease in average taxi-out time (achieving the average unimpeded taxi-out time).

Abstract: We present initial design of NextGen flow corridors and provide examples of corridor building blocks. We discuss important factors to consider when designing the shape and dimensions of corridor building blocks. In addition, we use existing procedures for Area Navigation (RNAV) routes as a basis for developing operational procedures to implement flow corridor operations. Sample procedures are presented to cover the role of pilots, controllers, or other Air Navigation Service Provider (ANSP), as well as the capabilities and information each party has to receive to perform the described procedures. Finally, required displayed functions for pilot and controller are discussed and illustrated in the context of selected scenarios. The work we present in this article is intended to be used for design and development of Human-In-The-Loop studies for corridors proof of concept.

Abstract: Scanning two book pages at the same time helps to accelerate the scanning process but on the other hand introduces several difficulties if the user needs to have one page per image. A major difficulty is the appearance of noisy black borders around text areas as well as of noisy black stripes between the two pages. In this paper, we propose a novel algorithm for detecting the page frames on double page document images. Our aim is to split the image into the two pages as well as to remove noisy borders. First we apply a pre-processing which includes binarization, noise removal and image smoothing. Then, we detect the vertical zones of the two pages. In this stage, we introduce the vertical white run projections which have been proved efficient for detecting vertical zones of text areas. Finally, the horizontal zones of the two pages are detected based on horizontal white run projections. The experimental results on several double page document images from fifteen different books demonstrate the effectiveness of the proposed technique.

Abstract: This paper proposes an opening recognition corpus, HIT-OR3C, and its construction toolkit to facilitate the unconstrained online Chinese handwriting text recognition. The characters of HIT-OR3C are collected through handwriting pad and are recorded and labeled automatically via the proposed handwriting document collection software OR3C Toolkit. HIT-OR3C consists of 5 subsets, namely GB1, GB2, Letter, Digit and Document. The first 4 corpora contain 6,825 categories produced by 122 persons and 832,650 samples in total. The document corpus is corresponding to 10 news articles that contain 2,442 categories produced by 20 persons and 77,168 samples in total. HIT-OR3C can be used for training and evaluation of character recognition algorithms. The OR3C Toolkit provides an efficient, device-independent, and unconstrained platform for the building of large scale handwriting corpus.

Abstract: Model-driven development (MDD) has become a key technique in systems and software engineering, including the aeronautic domain. It facilitates on systematic use of models from a very early phase of the design process and through various model transformation steps (semi-)automatically generates source code and documentation. However, on one hand, the use of model-driven approaches for the development of configuration data is not as widely used as for source code synthesis. On the other hand, we believe that, particular systems that make heavy use of configuration tables like the ARINC 653 standard can benefit from model-driven design by (i) automating error-prone configuration file editing and (ii) using model based validation for early error detection. In this paper, we will present the results of the European project DIANA that investigated the use of MDD in the context of Integrated Modular Avionics (IMA) and the ARINC 653 standard. In the scope of the project, a tool chain was implemented that generates ARINC 653 configuration tables from high-level architecture models. The tool chain was integrated with different target systems (VxWorks 653, SIMA) and evaluated during case studies with real-world and real-sized avionics applications.

Abstract: Repetition of layout structure is prevalent in document images. In document design, such repetition conveys the underlying logical and functional structure of the data. For example, in invoices, the names, unit prices, quantities and other descriptors of every line item are laid out in a consistent spatial structure. We propose a general method for extracting such repeated structure from documents. After receiving a single example of the structure to be found, the proposed method localizes additional instances of this structure in the same document and in additional documents. A wide variety of perceptually motivated cues (such as alignment and saliency) is used for this purpose. These cues are combined in a probabilistic model, and a novel algorithm for exact inference in this model is proposed and used. We demonstrate that this method can cope with complex instances of repeated structure and generalizes successfully across a wide range of structure variations.