Wei Jia-Quan, Wang Xiao-Yun, Zheng Xia-Lin, Tong Xin
Guangxi Key Laboratory of Agro-Environment and Agric-Products Safety, National Demonstration Center for Experimental Plant Science Education, College of Agriculture, Guangxi University, Nanning 530004, China.
Insects. 2024 Oct 30;15(11):849. doi: 10.3390/insects15110849.
Sound is an important medium of communication among insects. Some longhorn beetles produce sounds during their daily activities, and these sounds play a role in courtship, predation, and defense. However, whether there are differences in the sounds emitted by longhorn beetles and how to distinguish and recognize these sounds have not been investigated in detail. Here, the sounds of (Fabricius), (Pascoe), and (Pascoe) were collected, and the differences in their stridulatory organs were observed and compared using scanning electron microscopy (SEM). The characteristics of their sounds were analyzed using MATLAB. Linear prediction cepstral coefficients (LPCC) and Mel frequency cepstral coefficients (MFCC) were used to extract the sound features, and the support vector machine (SVM) model was used to identify the sounds of three species. The results showed that the stridulatory organs of three species of longhorn beetles differed in morphology and time domain, and the combination of MFCC and SVM had a better recognition ability. The difference in the stridulatory organs of longhorn beetles may be an important reason for the differences in the sounds they produce, and we discussed the application of insect sounds in insect classification.
声音是昆虫之间重要的交流媒介。一些天牛在日常活动中会发出声音，这些声音在求偶、捕食和防御中发挥作用。然而，天牛发出的声音是否存在差异以及如何区分和识别这些声音尚未得到详细研究。在此，收集了光肩星天牛（黄斑星天牛）（Fabricius）、云斑白条天牛（Pascoe）和桑天牛（Pascoe）的声音，并使用扫描电子显微镜（SEM）观察和比较了它们发声器官的差异。使用MATLAB分析了它们声音的特征。利用线性预测倒谱系数（LPCC）和梅尔频率倒谱系数（MFCC）提取声音特征，并使用支持向量机（SVM）模型识别三种天牛的声音。结果表明，三种天牛的发声器官在形态和时域上存在差异，MFCC和SVM的组合具有较好的识别能力。天牛发声器官的差异可能是它们发出声音不同的重要原因，并且我们讨论了昆虫声音在昆虫分类中的应用。
