Matko Šarić

Scene text segmentation using low variation extremal regions and sorting based character grouping

Extraction of textual information from natural scene images is a challenging task due to imaging conditions and diversity of text properties. Segmentation of scene text is important step in the pipeline that significantly affects the final recognition performance. In this paper I propose a new scene text segmentation method. Firstly, a novel approach for character candidates generation based on extremal regions (ERs) is introduced. Subpaths having low area variation are extracted from ER tree. Instead of using minimum variation criterion for selection of character candidates, position of ER in extracted subpath is used as criterion for that purpose. Each subpath is represented by one ER that is sent to SVM-based classification step. After that a novel method for character candidates grouping is used to discard non-character objects that are wrongly classified as characters. Proposed approach estimates vertical positions of the lines by sorting y coordinates of region centroids and checks spatial relation of adjacent regions in the line. This step enhances precision significantly and has lower computational complexity compared to hierarchical clustering methods. Finally, the last step is restoration of character ERs erroneously eliminated by SVM classifier where text layout properties are exploited to correct false negative classifications. Experimental results obtained on the ICDAR 2013 dataset show that the proposed character candidates generation method efficiently prunes repeating regions and achieves character recall rate superior to recently published ER based method. Proposed segmentation algorithm obtains competitive performance compared to state-of-the-art methods.

Improved Linearized Combinatorial Model (ILCM) for optimal frame size selection in ALOHA-based RFID systems

Radio Frequency Identification (RFID) technology became the most important tool for identification of items and tracking. Nowadays, the most popular in terms of best price-performance ratio is passive RFID technology, where tags are both powered-up and communicating using the same radio waves transmitted via reader antenna(s).

Cochlea-based Features for Music Emotion Classification

Listening to music often evokes strong emotions. With the rapid growth of easily-accessible digital music libraries there is an increasing need in reliable music emotion recognition systems. Common musical features like tempo, mode, pitch, clarity, etc. which can be easily calculated from audio signal are associated with particular emotions and are often used in emotion detection systems. Based on the idea that humans don’t detect emotions from pure audio signal but from a signal that had been previously processed by the cochlea, in this work we propose new cochlear based features for music emotion recognition. Features are calculated from the gammatone filterbank model output and emotion classification is then performed using Support Vector Machine (SVM) and TreeBagger classifiers. Proposed features are evaluated on publicly available 1000 songs database and compared to other commonly used features. Results show that our approach is effective and outperforms other commonly used features. In the combined features set we achieved accuracy of 83.88% and 75.12% for arousal and valence.