Recognition of Mexican Sign Language from Frames in Video Sequences
The development of vision systems capable to extracting discriminative features that enhance the generalization power of a classifier is still a very challenging problem. In this paper, is presented a methodology to improve the classification performance
- PDF / 1,023,939 Bytes
- 10 Pages / 439.37 x 666.142 pts Page_size
- 93 Downloads / 198 Views
sgrado e Investigacion, Autonomous University of Mexico State, 56259 Texcoco, Mexico [email protected] 2 Division of Research and Postgraduate Studies, Orizaba Technology Institute, Orizaba, Veracruz, Mexico 3 Faculty of Computer Science, Autonomous University of Sinaloa, Culiacán, Mexico 4 Centro Universitario UAEM-Zumpango, Camino Viejo a Jilotzingo continuación calle Rayón, 55600 Zumpango, Mexico
Abstract. The development of vision systems capable to extracting discriminative features that enhance the generalization power of a classifier is still a very challenging problem. In this paper, is presented a methodology to improve the classification performance of Mexican Sign Language (MSL). The proposed method explores some frames in video sequences for each sign. 743 features were extracted from these frames, and a genetic algorithm is employed to select a subset of sensitive features by removing the irrelevant features. The genetic algorithm permits to obtain the most discriminative features. Support Vector Machines (SVM) are used to classify signs based on these features. The experiments show that the proposed method can be successfully used to recognize the MSL with accuracy results individually above 97 % on average. The proposed feature extraction methodology and the GA used to extract the most discriminative features is a promising method to facilitate the communication of deaf people. Keywords: Mexican Sign Language
Classification Features
1 Introduction The development of devices that help to the people with limitations such as diminution of their physical abilities, intellectual or sensory is a current research in many fields. The development of tools to facilitate the communication using the Mexican Sign Language (MSL) between deaf people and hearing people is vital to facilitate interaction, dialogue and information in social and private scopes, in addition to providing access to education and employment. The development of tools to help the overall
© Springer International Publishing Switzerland 2016 D.-S. Huang and K.-H. Jo (Eds.): ICIC 2016, Part II, LNCS 9772, pp. 353–362, 2016. DOI: 10.1007/978-3-319-42294-7_31
354
J. Cervantes et al.
communication between transmitter (deaf people)-receiver [3, 7, 15], and transmitter-receiver (deaf people) [1, 2, 5, 6, 11, 12] is a major challenge of current research. The actual research is focused in both directions, transmitter-receiver (deaf people) and transmitter (deaf people)-receiver. The first one uses animated characters or avatars in video in which sign language interpreters translate the text. The second one tries to translate a dialogue or identify words in sign language issued by deaf people and translate it by vision systems. To translate a sign language dialogue is not an easy challenge because the signs used in the MSL are signs in movement or static signs at some point in the body or space. Each signal may be composed of five or even seven features; hand configuration, location, direction, orientation, non-manual features, place of articula
Data Loading...
