Search

In this paper we describe an experimental investigation into the significance of different facial parts in the task of age estimation. For this purpose we use an automatic age estimation algorithm capable of producing an estimate of the age of a subject based on his/her face image. We perform automatic age estimation experiments using different facial parts of varying resolution so that the significance of facial parts in this application can be assessed. According to the results obtained, the internal part of the face and in particular the area around the eyes proved to be the most significant for the task of automatic age estimation.

Summary form only given. This paper aims at an integrative and innovative approach to curriculum methods of teaching engineering methodology courses. The course is problem driven where industrial case studies are the basis for the introduction of problems, key concepts, and solution methodologies. The objective of developing this educational environment is to bring the complexity of real world decision making to the classroom and to simulate the dynamics and uncertainties of organisational decision making in a classroom setting.

In this paper an experimental X-band radar system, called DTU-VISDAM, developed for the detection and monitoring of human vital signs is described. The DTU-VISDAM radar exploits a low intermediate frequency (IF) heterodyne RF front-end architecture and single-sideband (SSB) transmission for easier and more reliable extraction of the vital signs. The hardware implementation of the proposed low-IF RF front-end architecture and associated IF circuitry is discussed. Furthermore, the signal processing and calibration steps necessary to extract the vital signs information measured on a human subject are described.

The safety-critical Java (SCJ) specification provides a restricted set of the Java language intended for applications that require certification. In order to test the specification, implementations are emerging and the need to evaluate those implementations in a systematic way is becoming important. In this paper we evaluate our SCJ implementation which is based on the Java Optimized Processor JOP and we measure different performance and timeliness criteria relevant to hard real-time systems. Our implementation targets Level 0 and Level1 of the specification and to test it we use a series of micro benchmarks, an application-based benchmark, and a reduced set of a SCJ technology compatibility kit. We evaluate the accuracy of periods, linear-time memory allocation, aperiodicevent handling, dispatch latency for interrupts, context switch preemption latency, and synchronization.

Non-negative Tensor Factorization (NTF) has become a prominent tool for analyzing high dimensional multi-way structured data. In this paper we set out to analyze gene expression across brain regions in multiple subjects based on data from the Allen Human Brain Atlas [1] with more than 40 % data missing in our problem. Our analysis is based on the non-negativity constrained Canonical Polyadic (CP) decomposition where we handle the missing data using marginalization considering three prominent alternating least squares procedures; multiplicative updates, column-wise, and row-wise updating of the component matrices. We examine three gene expression prediction scenarios based on data missing at random, whole genes missing and whole areas missing within a subject. We find that the column-wise updating approach also known as HALS performs the most efficient when fitting the model. We further observe that the non-negativity constrained CP model is able to predict gene expressions better than predicting by the subject average when data is missing at random. When whole genes and whole areas are missing it is in general better to predict by subject averages. However, we find that when whole genes are missing from all subjects the model based predictions are useful. When analyzing the structure of the components derived for one of the best predicting model orders the components identified in general constitute localized regions of the brain. Non-negative tensor factorization based on marginalization thus forms a promising framework for imputing missing values and characterizing gene expression in the human brain. However, care also has to be taken in particular when predicting the genetic expression levels at a whole region of the brain missing as our analysis indicates that this requires a substantial amount of subjects with data for this region in order for the model predictions to be reliable.

Experiences from a new course in mechatronics at Technical University of Denmark are conveyed in this paper. The course is supposed to teach students enrolled in the bachelor degree in electrical engineering some fundamental knowledge about mechanics and to teach students enrolled in the bachelor degree in mechanical engineering fundamentals about electronics. Furthermore the course uses project work as a method to keep the students actively participating and in part have them teach each other the subjects. The general course plan is presented and the reasoning behind the course structure is discussed in the paper, together with a brief look at the student's reactions in form of data from the course evaluation.

As research and development of renewable and clean energy based systems is advancing rapidly, the nonlinear source emulator (NSE) is becoming very essential for testing of maximum power point trackers or downstream converters. Renewable and clean energy sources play important roles in both terrestrial and nonterrestrial applications. However, most existing NSEs have only been concerned with simulating energy sources in terrestrial applications, which may not be fast enough for testing of nonterrestrial applications. In this paper, a high-bandwidth NSE is developed that is able to simulate the behaviors of a typical nonlinear source under different critical conditions that can happen during their operations. The proposed 200-W NSE, which consists of a fourth-order output filter buck converter and a novel nonlinear small-signal reference generator, can quickly react not only to an instantaneous change in the input source but also to a load step between nominal and open circuit. Moreover, all of these operation modes have a very fast settling time of only 10 $\mu$ s, which is hundreds of times faster than that of existing works. This attribute allows for higher speed and a more efficient maximum power point tracking algorithm. The proposed NSE, therefore, offers a superior dynamic performance among devices of the same kind.

Biomass steam pretreatment is a key process in converting agricultural wastes to bioethanol. The pretreatment occurs in a large pressurized tank called a thermal reactor. Two key parameters influence the successfulness of the process: the reactor temperature, and the retention time. A particle pump pressurizes untreated biomass from atmospheric to reactor pressure with recycled steam from the reactor. This paper formulates a steam mathematical model both for the thermal reactor and the particle pump, which is then used to design an L1 adaptive output feedback controller for the reactor temperature. As steam is recycled from the reactor to pressurize the particle pump, pressure drops and the reactor temperature is disturbed. The main control challenge is to reject these disturbances and keep a steady temperature. The nonlinear process model embeds mass and energy balances, valve characteristics, and enthalpy-pressure and pressure-temperature dependencies. Nonlinear feed-forward terms are added in the control strategy. The process model, the control strategy, the application of the L1 adaptive controller and its tuning method based on minimizing a cost function represent novelties of this paper.