The paper presents the analysis and discussion of gender recognition based on human face picture. The research combines different features selection techniques with the set of softcomputing classifiers. We are looking for not very complicated, fast and sensitive approach to create the theoretical basis for real safety systems where the correct “on-line” gender recognition is necessary. We start from the already known differences between the female and male face. This is the key point to tune the preprocessing mechanisms. We propose the quite classic classifiers, but we focus on sensible correlation between the feature extraction and the actual classification. The significant set of the results are discussed and the best solutions are pointed. All tests were realised based on the well known base of face pictures with added set of our own collection. The proposed solution can be an essential tool for the monitoring systems, safety guards and systems to point the dangerous situations based on video data.
Gender recognition, across different races and regardless of age, is becoming an increasingly important technology in the domains of marketing, human-computer interaction and security. Most state-of-the-art systems consider either highly constrained conditions or relatively large databases. In either case, often not enough attention is paid to cross-racial age-invariant applications. This paper proposes a method of hybrid classification, which performs well even with a small training set. The design of the classifier enables the construction of reliable decision boundaries insensitive to an aging model as well as to race variation. For a training set consisting of one hundred images, the proposed method reached an accuracy level of 90%, whereas the best method known from the literature, tested under the restrictions imposed on the database, achieved only 78% accuracy.
Soft biometrics methods that involve gender, age and ethnicity are still developed. Face recognition methods often rely on gender recognition. The same applies to the methods reconstructing the faces or building 2D or 3D models of the faces. In the paper, we conduct study on different set of gender recognition methods and their mobile applications. We show the advantages and disadvantages of that methods and future challenges to the researches. In the previous papers, we examined a range variety of skin detection methods that help to spot the face in the images or video stream. On acquiring faces, we focus on gender recognition that will allow us to create pattern to build 2D and 3D automatic faces models from the images. That will result also in face recognition and authentication, also.
4
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
W artykule opisano autorski algorytm rozpoznawania płci na podstawie analizy obrazu twarzy osoby. Kolejne etapy działania algorytmu to obróbka wstępna obrazu, detekcja twarzy w obrazie, ekstrakcja i selekcja cech oraz klasyfikacja. W ramach eksperymentów określono wpływ wybranych parametrów opracowanego algorytmu na efektywność klasyfikacji oraz zaproponowano rozwiązania mające na celu poprawę jego skuteczności.
EN
The article describes the gender recognition algorithm based on facial image analysis. The subsequent stages of the algorithm are: image preprocessing, face detection, feature extraction/selection and classification. In the framework of experiments, the influence of selected algorithm parameters on the efficiency of classification was determined and proposals how to improve effectiveness of classification was described.
The paper presents the simple technique of speaker gender recognition that uses MFCC features typically applied in automatic speech recognition. Artificial neural network is used as a classifier. The speech signal is first divided into 20 ms frames. For each frame, Mel-Frequency Cepstral Coefficients are extracted and the created feature vector is provided into a neural network classifier, which individually classifies each frame as male or female sample. Finally, the whole utterance is classified by selecting the class, for which the sum of corresponding neural network outputs is greater. The advantage of the method is that it can be easily combined with speech recognition, because both processes (gender recognition and speech recognition) are based on the same features. This way, no additional logic and no extra computational power is needed to extract features necessary for gender recognition. The method was experimentally evaluated using speech samples in English and in Polish. The comparison with other methods described in literature based on other feature extraction methods shows the superiority of the proposed approach, especially in cases where the recognition is carried out in noisy environment or using poor audio equipment.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.