Markus Turtinen

View-based recognition of real-world textures

Abstract A new method for recognizing 3D textured surfaces is proposed. Textures are modeled with multiple histograms of micro-textons, instead of more macroscopic textons used in earlier studies. The micro-textons are extracted with the recently proposed multiresolution local binary pattern operator. Our approach has many advantages compared to the earlier approaches and provides the leading performance in the classification of Columbia–Utrecht database textures imaged under different viewpoints and illumination directions. It also provides very promising results in the classification of outdoor scene images. An approach for learning appearance models for view-based texture recognition using self-organization of feature distributions is also proposed. The method performs well in experiments. It can be used for quickly selecting model histograms and rejecting outliers, thus providing an efficient tool for vision system training even when the feature data has a large variability.

Real-time surface inspection by texture

In this paper a real-time surface inspection method based on texture features is introduced. The proposed approach is based on the Local Binary Pattern (LBP) texture operator and the Self-Organizing Map (SOM). A very fast software implementation of the LBP operator is presented. The SOM is used as a powerful classifier and visualizer. The efficiency of the method is empirically evaluated in two different problems including textures from the Outex database and from a paper inspection problem.

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.

Contextual Analysis of Textured Scene Images.

Classifying image regions into one of several pre-defined semantic categories is a typical image understanding problem. Different image regions and object types might have very similar color or texture characteristics making it difficult to categorize them. Without contextual information it is often impossible to find reasonable semantic labeling for outdoor images. In this paper, we combine an efficient SVM-based local classifier with the conditional random field framework to incorporate spatial contex information to the classification. The images are represented with powerful local texture features. Then a discriminative multiclass model for finding good labeling for the image is learned. The performance of the method was evaluated with two different datasets. The approach was also shown to be useful in more general image retrieval and annotation tasks based on classification.

Face Tracking for Spatially Aware Mobile User Interfaces

This paper introduces a new face tracking approach for controlling user interfaces in hand-held mobile devices. The proposed method detects the face and the eyes of the user by employing a method based on local texture features and boosting. An extended Kalman filter combines local motion features extracted from the face region and the detected eye positions to estimate the 3-D position and orientation of the camera with respect to the face. The camera position is used as an input for the spatially aware user interface. Experimental results on real image sequences captured with a camera-equipped mobile phone validate the feasibility of the method.

Texture classification by combining local binary pattern features and a self-organizing map

This paper deals with the combined use of Local Binary Pattern (LBP) features and a Self-Organizing Map (SOM) in texture classification. With this approach, the unsupervised learning and visualization capabilities of a SOM are utilized with highly efficient histogram-based texture features. In addition to the Euclidean distance normally used with a SOM, an information theoretic log-likelihood (cumlog) dissimilarity measure is also used for determining distances between feature histograms. The performance of the approach is empirically evaluated with two different data sets: (1) a texture-based visual inspection problem containing four very similar paper classes, and (2) classification of 24 different natural textures from the Outex database.

View-based recognition of 3D-textured surfaces

A new method for recognizing 3D-textured surfaces is proposed. Textures are modeled with multiple histograms of micro-textons, instead of the more macroscopic textons used in earlier studies. The micro-textons are extracted with a recently proposed multiresolution local binary pattern operator. Our approach has many advantages compared to the earlier approaches and provides the leading performance in the classification of Columbia-Utrecht database (CUReT) textures imaged under different viewpoints and illumination directions. An approach for learning appearance models for view-based texture recognition using self-organization of feature distributions is also proposed.. It can be used for quickly selecting model histograms and rejecting outliers, thus providing an efficient tool for vision system training, even when the feature data has a large variability.

Texture-based paper characterization using nonsupervised clustering

A non-supervised clustering based method for classifying paper according to its quality is presented. The method is simple to train, requiring minimal human involvement. The approach is based on Self-Organizing Maps and texture features that discriminate the texture of effectively. Multidimensional texture feature vectors are first extracted from paper images. The dimensionality of the data is then reduced by a Self-Organizing Map (SOM). In dimensionality reduction, the feature data are projected to a two-dimensional space and clustered according to their similarity. The clusters represent different paper qualities and can be labeled according to the quality information of the training samples. After that, it is easy to find the quality class of the inspected paper by checking where a sample is placed in the low-dimensional space. Tests based on images taken in a laboratory environment from four different paper quality classes provided very promising results. Local Binary Pattern (LBP) texture features combined with a SOM-based approach classified the test data almost perfectly: the error percentage was only 0.2% with the multiresolution version of LBP and 1.6% with the regular LBP. The improvement to the previously used texture features in paper inspection is huge: the classification error is reduced over 40 times. In addition to the excellent classification accuracy, the method also offers a self-intuitive user interface and a synthetic view to the inspected data.

New video applications on mobile communication devices

The video applications on mobile communication devices have usually been designed for content creation, access, and playback. For instance, many recent mobile devices replicate the functionalities of portable video cameras and video recorders, and digital TV receivers. These are all demanding uses, but nothing new from the consumer point of view. However, many of the current devices have two cameras built in, one for capturing high resolution images, and the other for lower, typically VGA (640x480 pixels) resolution video telephony. We employ video to enable new applications and describe four actual solutions implemented on mobile communication devices. The first one is a real-time motion based user interface that can be used for browsing large images or documents such as maps on small screens. The motion information is extracted from the image sequence captured by the camera. The second solution is a real-time panorama builder, while the third one assembles document panoramas, both from individual video frames. The fourth solution is a real-time face and eye detector. It provides another type of foundation for motion based user interfaces as knowledge of presence and motion of a human faces in the view of the camera can be a powerful application enabler.

Performance evaluation of image noise reduction computing on a mobile platform

Noise reduction is one of the most fundamental digital image processing challenges. On mobile devices, proper solutions for this task can significantly increase the output image quality making the use of a camera even more attractive for customers. The main challenge is that the processing time and energy efficiency must be optimized, since the response time and the battery life are critical factors for all mobile applications. To identify the solutions that maximizes the real-time performance, we compare several different implementations in terms of computational performance and energy efficiency. Specifically, we compare the OpenCL based design with multithreaded and NEON accelerated implementations and analyze them on the mobile platform. Based on the results of this study, the OpenCL framework provides a viable energy efficient alternative for implementing computer vision algorithms.