Olli Silvén

A four-step camera calibration procedure with implicit image correction

In geometrical camera calibration the objective is to determine a set of camera parameters that describe the mapping between 3-D reference coordinates and 2-D image coordinates. Various methods for camera calibration can be found from the literature. However surprisingly little attention has been paid to the whole calibration procedure, i.e., control point extraction from images, model fitting, image correction, and errors originating in these stages. The main interest has been in model fitting, although the other stages are also important. In this paper we present a four-step calibration procedure that is an extension to the two-step method. There is an additional step to compensate for distortion caused by circular features, and a step for correcting the distorted image coordinates. The image correction is performed with an empirical inverse model that accurately compensates for radial and tangential distortions. Finally, a linear method for solving the parameters of the inverse model is presented.

A real-time system for monitoring of cyclists and pedestrians

Camera based fixed systems are routinely used for monitoring highway traffic. For this purpose inductive loops and microwave sensors are mainly used. Both techniques achieve very good counting accuracy and are capable of discriminating trucks and cars. However pedestrians and cyclists are mostly counted manually. In this paper, we describe a new camera based automatic system that utilizes Kalman filtering in tracking and Learning Vector Quantization (LVQ) for classifying the observations to pedestrians and cyclists. Both the requirements for such systems and the algorithms used are described. The tests performed show that the system achieves around 80%-90% accuracy in counting and classification.

Calibration procedure for short focal length off-the-shelf CCD cameras

A camera calibration procedure intended for a 3D measurement application is presented, paying attention to the various error sources. The error may be measurement noise that is random by nature, but it may also be systematic originating from the calibration target used, geometrical distortions and illumination. In order to obtain good calibration results, the systematic error sources should be eliminated or their effects compensated for. Then, the camera parameters can be determined by fitting the corrected measurements to the camera model which in our case is a combination of a pinhole camera and lens distortion models. We also notice that a more complete camera model is needed to explain all the error components.

Wood inspection with non-supervised clustering

Abstract. The appearance of sawn timber has huge natural variations that the human inspector easily compensates for mentally when determining the types of defects and the grade of each board. However, for automatic wood inspection systems these variations are a major source for complication. This makes it difficult to use textbook methodologies for visual inspection. These methodologies generally aim at systems that are trained in a supervised manner with samples of defects and good material, but selecting and labeling the samples is an error-prone process that limits the accuracy that can be achieved. We present a non-supervised clustering-based approach for detecting and recognizing defects in lumber boards. A key idea is to employ a self-organizing map (SOM) for discriminating between sound wood and defects. Human involvement needed for training is minimal. The approach has been tested with color images of lumber boards, and the achieved false detection and error escape rates are low. The approach also provides a self-intuitive visual user interface.

Face and Eye Detection for Person Authentication in Mobile Phones

Computer vision applications for mobile phones are gaining increasing attention due to several practical needs resulting from the popularity of digital cameras in todays mobile phones. In this work, we consider the task of face detection and authentication in mobile phones and experimentally analyze a face authentication scheme using Haar-like features with Ad-aBoost for face and eye detection, and local binary pattern (LBP) approach for face authentication. For comparison, another approach to face detection using skin color for fast processing is also considered and implemented. Despite the limited CPU and memory capabilities of todays mobile phones, our experimental results show good face detection performance and average authentication rates of 82% for small-sized faces (40times40 pixels) and 96% for faces of 80times80 pixels. The system is running at 2 frames per second for images of 320times240 pixels. The obtained results are very promising and assess the feasibility of face authentication in mobile phones. Directions for further enhancing the performance of the system are also discussed.

Accelerating image recognition on mobile devices using GPGPU

The future multi-modal user interfaces of battery-powered mobile devices are expected to require computationally costly image analysis techniques. The use of Graphic Processing Units for computing is very well suited for parallel processing and the addition of programmable stages and high precision arithmetic provide for opportunities to implement energy-efficient complete algorithms. At the moment the first mobile graphics accelerators with programmable pipelines are available, enabling the GPGPU implementation of several image processing algorithms. In this context, we consider a face tracking approach that uses efficient gray-scale invariant texture features and boosting. The solution is based on the Local Binary Pattern (LBP) features and makes use of the GPU on the pre-processing and feature extraction phase. We have implemented a series of image processing techniques in the shader language of OpenGL ES 2.0, compiled them for a mobile graphics processing unit and performed tests on a mobile application processor platform (OMAP3530). In our contribution, we describe the challenges of designing on a mobile platform, present the performance achieved and provide measurement results for the actual power consumption in comparison to using the CPU (ARM) on the same platform.

Visual Characterization of Paper Using Isomap and Local Binary Patterns

In this paper, we study how a multidimensional local binary pattern (LBP) texture feature data can be visually explored and analyzed. The goal is to determine how true paper properties can be characterized with local texture features from visible light images. We utilize isometric feature mapping (Isomap) for the LBP texture feature data and perform non-linear dimensionality reduction for the data. These 2D projections are then visualized with original images to study data properties. Visualization is utilized in the manner of selecting texture models for unlabeled data and analyzing feature performance when building a training set for a classifier. The approach is experimented on with simulated image data illustrating different paper properties and on-line transilluminated paper images taken from a running paper web in the paper mill. The simulated image set is used to acquire quantitative figures on the performance while the analysis of real-world data is an example of semi-supervised learning.

Comparison of dimensionality reduction methods for wood surface inspection

Dimensionality reduction methods for visualization map the original high-dimensional data typically into two dimensions. Mapping preserves the important information of the data, and in order to be useful, fulfils the needs of a human observer. We have proposed a self-organizing map (SOM)- based approach for visual surface inspection. The method provides the advantages of unsupervised learning and an intuitive user interface that allows one to very easily set and tune the class boundaries based on observations made on visualization, for example, to adapt to changing conditions or material. There are, however, some problems with a SOM. It does not address the true distances between data, and it has a tendency to ignore rare samples in the training set at the expense of more accurate representation of common samples. In this paper, some alternative methods for a SOM are evaluated. These methods, PCA, MDS, LLE, ISOMAP, and GTM, are used to reduce dimensionality in order to visualize the data. Their principal differences are discussed and performances quantitatively evaluated in a few special classification cases, such as in wood inspection using centile features. For the test material experimented with, SOM and GTM outperform the others when classification performance is considered. For data mining kinds of applications, ISOMAP and LLE appear to be more promising methods.

Effects of repeatability measures on results of fMRI sICA: A study on simulated and real resting-state effects

Spatial independent components analysis (sICA) has become a widely applied data-driven method for fMRI data, especially for resting-state studies. These sICA approaches are often based on iterative estimation algorithms and there are concerns about accuracy due to noise. Repeatability measures such as ICASSO, RAICAR and ARABICA have been introduced as remedies but information on their effects on estimates is limited. The contribution of this study was to provide more of such information and test if the repeatability analyses are necessary. We compared FastICA-based ordinary and repeatability approaches concerning mixing vector estimates. Comparisons included original FastICA, FSL4 Melodic FastICA and original and modified ICASSO. The effects of bootstrapping and convergence threshold were evaluated. The results show that there is only moderate improvement due to repeatability measures and only in the bootstrapping case. Bootstrapping attenuated power from time courses of resting-state network related ICs at frequencies higher than 0.1 Hz and made subsets of low frequency oscillations more emphasized IC-wise. The convergence threshold did not have a significant role concerning the accuracy of estimates. The performance results suggest that repeatability measures or strict converge criteria might not be needed in sICA analyses of fMRI data. Consequently, the results in existing sICA fMRI literature are probably valid in this sense. A decreased accuracy of original bootstrapping ICASSO was observed and corrected by using centrotype mixing estimates but the results warrant for thorough evaluations of data-driven methods in general. Also, given the fMRI-specific considerations, further development of sICA methods is strongly encouraged.

Fixed- and Floating-Point Processor Comparison for MIMO-OFDM Detector

The evolution toward software-defined radio (SDR) technologies, in particular, cognitive radios, is leading toward the need to support multiple radio solutions with the same baseband processing resources. This implies not only a huge design effort, but also a shift from hardware to software design flavored tool chains. In this paper, a hardware complexity and energy dissipation are analyzed by implementing three programmable processor architectures that support 32- and 12-bit floating-point and 16-bit fixed-point arithmetics. The processors are based on the transport triggered architecture (TTA) that has a very low programmability overhead. We programmed a recently introduced selective spanning with fast enumeration (SSFE) soft-output detector for these processors. The processors are capable to achieve data rates required in multiple-input multiple-output orthogonal frequency- division multiplexing (MIMO-OFDM) 3G LTE system with a small energy dissipation. The analysis shows that at the same goodput rate a floating-point implementation can achieve a lower gate count and a better power efficiency than a fixed-point design. Combined with tool chain benefits, the floating-point arithmetic is becoming attractive for future SDR solutions.