Abstract: While computer-based design tools are widely used in architecure during late design phases for creating final floor plans, early design phases usually still take place in a traditional manner, using pen, paper and scissors. At the beginning of these phases, there is often only a rough idea of how a building should look like. Viewing existing floorplans of similar buildings can help an architect in his/her creative work, but searching for those plans manually is very time-consuming. Automated tools for searching similar floor plans could help to lower the amount of time needed for such investigations tremendously. In order to employ such search mechanisms, proper user interfaces are needed that fit to the architect’s working process. These interfaces should be useable easily and naturally, requiring less initial training. They should be capable of creating search requests that can be processed by the attached search mechanism. In this article, we describe two different user interfaces to serve this purpose. We describe their structures and interaction principles. Afterwards we show their general usability and user acceptance by the means of a users study.

Abstract: It is very common to find different versions of the same music work in archives of Opera Theaters. These differences correspond to modifications and annotations from the musicians. From the musicologist point of view, these variations are very interesting and deserve study. This paper explores the alignment of music scores as a tool for automatically detecting the passages that contain such differences. Given the difficulties in the recognition of handwritten music scores, our goal is to align the music scores and at the same time, avoid the recognition of music elements as much as possible. After removing the staff lines, braces and ties, the bar lines are detected. Then, the bar units are described as a whole using the Blurred Shape Model. The bar units alignment is performed by using Dynamic Time Warping. The analysis of the alignment path is used to detect the variations in the music scores. The method has been evaluated on a subset of the CVC-MUSCIMA dataset, showing encouraging results.

Abstract: Circle and circular arc detection in images have been a long standing topic in image analysis. It finds numerous applications for both scanned document images as well as in photographic images. As a result, circle detection algorithms are published regularly and benchmarking data sets and contests have been organized on a regular basis over the last decades. Unfortunately, they have not been able to achieve a very clear image establishing which approaches perform best and under what exact conditions.

Abstract: This paper addresses performance evaluation in the presence of imprecise ground-truth. Indeed, the most common assumption when performing benchmarking measures is that the reference data is flawless. In previous work, we have shown that this assumption cannot be taken for granted, and that, in the case of perceptual interpretation problems it is most certainly always wrong but for the most trivial cases.

Abstract: The aim of this paper is to develop a set of algorithms for the automated construction of synthetic digital ground truth images from historic cosmic ray recordings. These images can subsequently be used to test data extraction algorithms. This takes place in a larger research context of an effort to retrieve and digitize the data contained within more than 20 years (1934–1956) of historic cosmic ray data from around the world. The creation of synthetic ground truth images can logically be broken down into component tasks, which can be approached individually. These tasks include: binarization, segmentation, as well as generation of optical artefacts and distortions. The approach and details of the algorithm are described.

Abstract: With the objective of creating an interface for experimenting with electronic circuits embedded in documents or images, in this paper we have presented a system for parsing and understanding of electronic circuit diagrams. The developed system consists of following steps- symbol extraction, symbol recognition, optimization and netlist-representation. Firstly, symbols are extracted from the image by removing text and connection lines using computer vision techniques. For symbol recognizer a probabilistic-SVM classifier is built using HOG and radon features on training data. A Bayesian framework is used to incorporate domain knowledge information to improve the performance of the probabilistic symbol recognizer. An novel optimization approach based on top-down features is used to remove the errors that occurs in the symbol extraction and recognition task. A depth first traversal algorithm is used to find the connections between the symbols and then image is represented in the form of usable data structure. The system is evaluated on a dataset of 20 analog electronic circuit images collected from various sources and the results are presented.

Abstract: The present paper proposes a technique for text- graphics separation of geographical maps based on vectorization process and decision tree classification. In the proposed method, every map image is vectorized in order to extract a set of features for characterizing text and graphics. Vectorization provides structural primitives. We associate features to these structural primitives. A decision tree is then designed to discriminate text and graphics in map images, considering the features extracted from the vectorized images. This method provides a binary decision for every vectorized component, classifying the components into graphic or text. The proposed method was tested on a Bangla (a popular Indian regional language) maps dataset composed of a set of grey level images. The proposed text- graphic separation method provides 72.6% and 67.01% character and word-level text extraction accuracy respectively, when tested on map images.

Abstract: In this paper, we propose a camera-based document retrieval system using various local features as well as indexing methods. For feature extraction, we use well known features such as LLAH, SIFT, SURF and ORB that are invariant to image transformations and work well with images captured by cameras. In addition, we employ our new features, named as Scale and Rotation Invariant Features (SRIF). SRIF is computed based on geometrical constraints between pairs of nearest points around a keypoint. Our systems are applied on dataset including 400 heterogeneous-content complex linguistic map images (huge size, 9800\(\,\times \,\)11768 pixels resolution). The experimental results show that the system using SRIF is efficient in terms of retrieval time with 95.2% retrieval accuracy.

Abstract: The graphics recognition research community has been employing graphs, in one form or another, for at-least the last three decades. These data-structures have proven to be the most powerful representations for encoding the structural information of underlying data, for further processing. However, there is still a lack of tools and methods which could be employed to process these useful data-structures in an efficient manner. Graph embedding provides a solution for this problem. In this paper we present an improvement of the Fuzzy Multilevel Graph Embedding (FMGE) technique, by adding new topological node features, named Morgan Index. The experimental results on GREC, Mutagenicity and Fingerprint datasets from IAM graph database, illustrate improved results for the graph classification and graph clustering problems.

Abstract: In this paper we present a knowledge base of architectural documents aiming at improving existing methods of floor plan classification and understanding. It consists of an ontological definition of the domain and the inclusion of real instances coming from both, automatically interpreted and manually labeled documents. The knowledge base has proven to be an effective tool to structure our knowledge and to easily maintain and upgrade it. Moreover, it is an appropriate means to automatically check the consistency of relational data and a convenient complement of hard-coded knowledge interpretation systems.