Robust cepstral feature for bird sound classification

Ramashini, Murugaiya; Abas, Pg Emeroylariffion; Mohanchandra, Kusuma; Silva, Liyanage C. De

doi:10.11591/ijece.v12i2.pp1477-1487

Cited by 15 publications

(4 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Commonly, a single metric, the F1 score, which combines image precision and recall [23,24], is used as a metric. F1 score is defined as the harmonic mean between precision and recall and may be determined using…”

Section: CIII Automated Cell Counting Of Histology Imagesmentioning

confidence: 99%

Automated Cell Counting using Image Processing

Hassan,

Suhaimi,

Bilad

et al. 2023

IJC

View full text Add to dashboard Cite

Manual cell counting using Hemocytometer is commonly used to quantify cells, as it is an inexpensive and versatile method. However, it is labour-intensive, tedious, and time-consuming. On the other hand, most automated cell counting methods are expensive and require experts to operate. Thus, the use of image analysis software allows one to access low-cost but robust automated cell counting. This study explores the advanced setting of image processing software to obtain routes with the highest counting accuracy. The results show the effectiveness of advanced settings in CellProfiler for counting cells from synthetic images. Two routes were found to give the highest performance, with average image and cell accuracies of 85% and 99.8%, respectively, and the highest F1 score of 0.83. However, the two routes were unable to correctly determine the exact number of cells on the histology images, albeit giving a respectable cell accuracy of 79.6%. Further investigation has shown that CellProfiler is able to correctly identify the bulk of the cells within the histology images. Good image quality with high focus and less blur was identified as the key to successful image-based cell counting. To further enhance the accuracy, other modules can be included to further segment an object hence improving the number of objects identified. Future work can focus on evaluating the robustness of the routes by comparing them with other methods and validating with the manual cell counting method.

show abstract

Section: CIII Automated Cell Counting Of Histology Imagesmentioning

confidence: 99%

Automated Cell Counting using Image Processing

Hassan,

Suhaimi,

Bilad

et al. 2023

IJC

View full text Add to dashboard Cite

show abstract

“…In recent years various researchers have proposed deep learning-based methods for activity sequence learning, which is inspired by the effectiveness of deep learning in applications, including video captioning [36], audio recognition [37], neural machine translation [38], image recognition [39], and speech recognition [40][41] [42]. Other deep learning approaches based on skeletal data include Recurrent Neural Network (RNN) [5][43], Convolutional Neural Network (CNN) [44], and Graph Convolutional Network (GCN) [45].…”

Section: Related Workmentioning

confidence: 99%

A Comparative Study of Supervised and Unsupervised Approaches in Human Activity Analysis Based on Skeleton Data

Hossen,

Abas

2023

IJCDS

View full text Add to dashboard Cite

One of the important areas of machine intelligence research today is human activity recognition (HAR), with the goal of automatically identifying human activities from various types of sensor data. Most of the existing human activity recognition methods use hand-crafted features and labelled data, but these methods fail to identify new activities not defined in the training dataset. As human activities are numerous and executed in various ways, it is challenging to obtain enough labelled data to train a model to recognize the different activities. In this paper, the performance of five different supervised learning algorithms on the human activity recognition task with skeleton-based features has been evaluated, using five publicly available datasets and an experimental dataset. Accuracies of above 90% are achievable on datasets with a limited number of samples using commonly available classification algorithms and simple skeleton-based features. Subsequently, the same feature sets are used on unsupervised learning methods for an unsupervised clustering task. Using the unsupervised learning algorithms, an average of 74% f1-score on the publicly available CAD60 dataset and 61% f1-score on the experimental dataset, are obtained. These results demonstrate the effectiveness of simple skeleton-based features, coupled with common supervised and unsupervised learning algorithms in human activity recognition tasks.

show abstract

“…Time windows of multiple durations and varying resolutions have been considered for the segmentation and feature representations. These features have been individually fed into a Support Vector Machine (SVM) classifier [29] for native language identification. Performances of the different features, effects of time window duration, and resolutions have been compared.…”

Section: Speech Features Analysismentioning

confidence: 99%

Speaker Profiling Based on the Short-Term Acoustic Features of Vowels

Humayun,

Shuja,

Abas

2023

Technologies

View full text Add to dashboard Cite

Speech samples can provide valuable information regarding speaker characteristics, including their social backgrounds. Accent variations with speaker backgrounds reflect corresponding acoustic features of speech, and these acoustic variations can be analyzed to assist in tracking down criminals from speech samples available as forensic evidence. Speech accent identification has recently received significant consideration in the speech forensics research community. However, most works have utilized long-term temporal modelling of acoustic features for accent classification and disregarded the stationary acoustic characteristics of particular phoneme articulations. This paper analyzes short-term acoustic features extracted from a central time window of English vowel speech segments for accent discrimination. Various feature computation techniques have been compared for the accent classification task. It has been found that using spectral features as an input gives better performance than using cepstral features, with the lower filters contributing more significantly to the classification task. Moreover, detailed analysis has been presented for time window durations and frequency bin resolution to compute short-term spectral features concerning accent discrimination. Using longer time durations generally requires higher frequency resolution to optimize classification performance. These results are significant, as they show the benefits of using spectral features for speaker profiling despite the popularity of cepstral features for other speech-related tasks.

show abstract

Robust cepstral feature for bird sound classification

Cited by 15 publications

References 27 publications

Automated Cell Counting using Image Processing

Automated Cell Counting using Image Processing

A Comparative Study of Supervised and Unsupervised Approaches in Human Activity Analysis Based on Skeleton Data

Speaker Profiling Based on the Short-Term Acoustic Features of Vowels

Contact Info

Product

Resources

About