Frog Identification System Based on Local Means K-Nearest Neighbors with Fuzzy Distance Weighting

Methodology for automatic bioacoustic classification of anurans based on feature fusion

“…In present research, this data collection of sound recordings has also been employed selecting 41 anuran species including those used in (Yuan & Ramli, 2013). Later, authors in (Jaafar et al, 2014) made another interesting comparative study on two databases with 13 and 15 frog species respectively, employing MFCC coefficients to train A C C E P T E D M A N U S C R I P T three classifiers: SVM, Sparse Representation Classifier (SRC) and Local Mean KNN with Fuzzy Distance Weighting (LMkNN-FDW). The experimental results of LMkNN-FDW provided the best result with 98.4% on the first database but only 87.2% on the second, due to calls could not be characterized successfully, with some species below 50%.…”

Fusion of Linear and Mel Frequency Cepstral Coefficients for Automatic Classification of Reptiles

“…MFCCs and LFCCs have been broadly used in speech recognition with success [21], and they have also been applied in animal bioacoustic classification [22][23][24][25], due to their easy implementation and high performance. Reptiles mainly produce sounds in low frequencies within the human auditory range.…”

Frog sound identification using extendedk-nearest neighbor classifier

“…Frog sound recognition gain a lot of interest among researcher and several techniques including syllable segmentation [1], feature extraction and classification [2] have been developed in order to recognize the frog species based on their sound. A typical frog sound recognition system consists of five major stages including data collection, data preprocessing to improve the quality of data, signal segmentation, feature extraction and classification [3]. The KNN introduced by [4] is a classical non-parametric classifier that attracts interest among researchers in pattern classification and recognition.…”

“…Several weighting schemes has been proposed to resolve the ambiguity of the weighting distance between the testing sample and its nearest neighbors. Empirical works on frog sound recognition with MFCC as feature extraction and various types of a classifier such as Support Vector Machine (SVM), Sparse Representation Classifier (SRC) and Local Mean k -Nearest Neighbor with Fuzzy Distance Weighting (LMkNN-FDW)has been proposed in [3]. The fuzziness introduced in LMkNN-FDW [3] is based on fuzzy set theories that assign the fuzzy membership to the testing sample [6].…”

“…Empirical works on frog sound recognition with MFCC as feature extraction and various types of a classifier such as Support Vector Machine (SVM), Sparse Representation Classifier (SRC) and Local Mean k -Nearest Neighbor with Fuzzy Distance Weighting (LMkNN-FDW)has been proposed in [3]. The fuzziness introduced in LMkNN-FDW [3] is based on fuzzy set theories that assign the fuzzy membership to the testing sample [6]. Another improvement of weighting scheme is proposed in [7] which the improved fuzzy is implemented with k -Nearest Centroid Neighbor (KNCN) classifier.…”