Find my office: Navigating real space from semantic descriptions

This paper shows that by using only symbolic language phrases, a mobile robot can purposefully navigate to specified rooms in previously unexplored environments. The robot intelligently organises a symbolic language description of the unseen environment and “imagines” a representative map, called the abstract map. The abstract map is an internal representation of the topological structure and spatial layout of symbolically defined locations. To perform goal-directed exploration, the abstract map creates a high-level semantic plan to reason about spaces beyond the robot’s known world. While completing the plan, the robot uses the metric guidance provided by a spatial layout, and grounded observations of door labels, to efficiently guide its navigation. The system is shown to complete exploration in unexplored spaces by travelling only 13.3% further than the optimal path.

Visual detection of occluded crop: For automated harvesting

This paper presents a novel crop detection system applied to the challenging task of field sweet pepper (capsicum) detection. The field-grown sweet pepper crop presents several challenges for robotic systems such as the high degree of occlusion and the fact that the crop can have a similar colour to the background (green on green). To overcome these issues, we propose a two-stage system that performs per-pixel segmentation followed by region detection. The output of the segmentation is used to search for highly probable regions and declares these to be sweet pepper. We propose the novel use of the local binary pattern (LBP) to perform crop segmentation. This feature improves the accuracy of crop segmentation from an AUC of 0.10, for previously proposed features, to 0.56. Using the LBP feature as the basis for our two-stage algorithm, we are able to detect 69.2% of field grown sweet peppers in three sites. This is an impressive result given that the average detection accuracy of people viewing the same colour imagery is 66.8%.

Place categorization and semantic mapping on a mobile robot

In this paper we focus on the challenging problem of place categorization and semantic mapping on a robot with-out environment-specific training. Motivated by their ongoing success in various visual recognition tasks, we build our system upon a state-of-the-art convolutional network. We overcome its closed-set limitations by complementing the network with a series of one-vs-all classifiers that can learn to recognize new semantic classes online. Prior domain knowledge is incorporated by embedding the classification system into a Bayesian filter framework that also ensures temporal coherence. We evaluate the classification accuracy of the system on a robot that maps a variety of places on our campus in real-time. We show how semantic information can boost robotic object detection performance and how the semantic map can be used to modulate the robot’s behaviour during navigation tasks. The system is made available to the community as a ROS module.

Personporträttet som tidningsgenre : En närläsningsstudie med fokus på innehållsliga teman, berättarkonventioner och kön

The thesis The portrait interview as a newspaper genre. A qualitative close reading focussing on topical motifs, conventions of narration, and gender defines the portrait interview as a newspaper genre and analyses how the personalities in the portraits are constructed textually. The main body of material consists of 107 portrait interviews in two morning newspapers, Dagens Nyheter (published in Stockholm, Sweden) and Hufvudstadsbladet (published in Swedish in Helsinki, Finland), during two one-week periods (week 46/1999 and week 38/2002). There is also complementary material of 59 portraits from four magazines. The study is carried out within the research traditions of journalistic genre studies, gender and journalism, and critical text analysis. It is comprised of a qualitative close reading focussing on content (topical motifs or themes), conventions of narration, and gender. The methods used to carry out the study are qualitative close reading and quantitative content analysis. The analysis identifies the stylistic elements that differentiate the portrait genre from other journalistic genres, as well as from the autobiographical genre, and explores what opportunities and limitations these elements present for the inclusion of even more women protagonists in the portrait genre. The portrait interview is an exception from the critical mission of journalism in general, with its position as a genre of politeness. Since a typical characteristic of the portrait interview genre is that it pays tribute to the protagonist, the genre reveals the kind of personalities and lives that are seen as admirable in society. Four levels of portrait interview are defined: the prototype portrait, the pure portrait, the hybrid portrait and the marginal portrait. The prototype is a raw version of a portrait that fulfils the criteria but may be lacking in content and stylistics. The pure portrait does not lack these qualities and resembles an ideal portrait. The hybrid is a borderline case which relates to another genre or is a mixture between the portrait and some other genre, most commonly the news genre. The marginal portrait does not fulfil the criteria, and can therefore be seen as an inadequate portrait. For example, obituaries and caricatures are excluded if the protagonist s voice is never quoted. The analysis resulted in three factors that in part help to explain why the portrait interview genre has somewhat more female protagonists than journalistic news texts do in general. The four main reasons why women are presented somewhat more in the portrait genre than in other journalistic genres are: (i) women are shown as exceptions to the female norm when, for example, taking a typical male job or managing in positions where there are few women; (ii) women are shown as representing female themes ; (iii) use of the double bind as a story-generating factor; and (iv) the intimisation of journalism. The double bind usually builds up the narration on female ambiguity in the contradiction between private and public life, for example family and career, personal desire and work. The intimisation of journalism and the double bind give women protagonists somewhat more publicity also because of the tendency of portrait interviews to create conflicts within the protagonist, as an exception to journalism in general where conflicts are created or seen as existing between, for example, persons, groupings or parties. Women protagonists and their lives create an optimal narration of inner conflicts originating in the double bind as men are usually not seen as suffering from these conflicts. The analysis also resulted in gendered portrait norms: The feminine portrait norm and the masculine portrait norm or more concretely, professional life and family life as expectation and exception. Women are expected to be responsible parents and mediocre professionals, while men are expected to be professionals and in their free time engaging fathers. Key words: journalism, genre, portrait interview, gender, interview, newspaper, women s magazine.

Determinant: Old algorithms, new insights (Extended Abstract)

In this paper we approach the problem of computing the characteristic polynomial of a matrix from the combinatorial viewpoint. We present several combinatorial characterizations of the coefficients of the characteristic polynomial, in terms of walks and closed walks of different kinds in the underlying graph. We develop algorithms based on these characterizations, and show that they tally with well-known algorithms arrived at independently from considerations in linear algebra.

Nano Al2O3-Pb AND SiO2-Pb cermets by sol-gel technique and the phase transformation study of the embedded Pb particles

Sol-gel processing followed by H2 reduction is used to produce dispersions of nanosized Pb in amorphous SiO2 and ultrafine γ Al2O3 matrices. A depression of 3–5K in Pb melting point is reported. The size and shape of these metastable particles in molten and solid state are discussed in the light of the experimental observations and expectations from the intersection group theory for equilibrium shape.

Lesion Detection in Breast Ultrasound Images Using Tissue Transition Analysis

Breast cancer is one of the leading cause of cancer related deaths in women and early detection is crucial for reducing mortality rates. In this paper, we present a novel and fully automated approach based on tissue transition analysis for lesion detection in breast ultrasound images. Every candidate pixel is classified as belonging to the lesion boundary, lesion interior or normal tissue based on its descriptor value. The tissue transitions are modeled using a Markov chain to estimate the likelihood of a candidate lesion region. Experimental evaluation on a clinical dataset of 135 images show that the proposed approach can achieve high sensitivity (95 %) with modest (3) false positives per image. The approach achieves very similar results (94 % for 3 false positives) on a completely different clinical dataset of 159 images without retraining, highlighting the robustness of the approach.

Average Overlap for Clustering Incomplete Data using Symmetric Non-Negative Matrix Factorization

Clustering techniques which can handle incomplete data have become increasingly important due to varied applications in marketing research, medical diagnosis and survey data analysis. Existing techniques cope up with missing values either by using data modification/imputation or by partial distance computation, often unreliable depending on the number of features available. In this paper, we propose a novel approach for clustering data with missing values, which performs the task by Symmetric Non-Negative Matrix Factorization (SNMF) of a complete pair-wise similarity matrix, computed from the given incomplete data. To accomplish this, we define a novel similarity measure based on Average Overlap similarity metric which can effectively handle missing values without modification of data. Further, the similarity measure is more reliable than partial distances and inherently possesses the properties required to perform SNMF. The experimental evaluation on real world datasets demonstrates that the proposed approach is efficient, scalable and shows significantly better performance compared to the existing techniques.