CnnPRL： progressive representation learning method for speech emotion recognition

doi:10.11772/j.issn.1001-9081.2024111704

Journal of Computer Applications ›› 2025, Vol. 45 ›› Issue (12): 3804-3812.DOI: 10.11772/j.issn.1001-9081.2024111704

• Artificial intelligence • Previous Articles Next Articles

CnnPRL： progressive representation learning method for speech emotion recognition

Yonghong FAN¹^,²^,³, Heming HUANG¹^,²^,³

^1.College of Computer，Qinghai Normal University，Xining Qinghai 810008，China
^2.The State Key Laboratory of Tibetan Intelligent Information Processing and Application （Qinghai Normal University），Xining Qinghai 810008，China
^3.Key Laboratory of Tibetan Information Processing，Ministry of Education （Qinghai Normal University），Xining Qinghai 810008，China

Received:2024-12-06 Revised:2025-03-08 Accepted:2025-03-18 Online:2025-03-21 Published:2025-12-10
Contact: Heming HUANG
About author:FAN Yonghong， born in 1997， Ph. D. candidate. Her research interests include pattern recognition， intelligent systems， speech emotion recognition.
HUANG Heming， born in 1969， Ph. D.， professor. His research interests include pattern recognition， intelligent systems， speech emotion recognition， speech enhancement， image processing.
Supported by:
National Natural Science Foundation of China(62066039);Natural Science Foundation of Qinghai Province(2022-ZJ-925);“111” Project(D20035)

渐进式表征学习语音情感识别方法CnnPRL

樊永红¹^,²^,³, 黄鹤鸣¹^,²^,³

^1.青海师范大学计算机学院，西宁 810008
^2.省部共建藏文智能信息处理及应用国家重点实验室（青海师范大学），西宁 810008
^3.藏文信息处理教育部重点实验室（青海师范大学），西宁 810008

通讯作者: 黄鹤鸣
作者简介:樊永红（1997—），女，宁夏吴忠人，博士研究生，CCF会员，主要研究方向：模式识别、智能系统、语音情感识别
黄鹤鸣（1969—），男（藏族），青海海东人，教授，博士，CCF会员，主要研究方向：模式识别、智能系统、语音情感识别、语音增强、图像处理。
基金资助:
国家自然科学基金资助项目(62066039);青海省自然科学基金资助项目(2022-ZJ-925);“111”项目(D20035)

Abstract

Abstract:

Speech Emotion Recognition （SER） aims to equip computers with the ability of identifying emotional states in speech signals accurately， and how to represent emotional features in speech efficiently is always a key research focus in SER. Currently， most research efforts are dedicated to leveraging deep learning methods to learn optimal features directly from raw speech or spectrogram data. This way is good at extracting more comprehensive representation information， but it may overlook the learning of more refined information within specific features and may not guarantee feature interpretability. Therefore， a Progressive Representation Learning method for SER based on Convolutional neural network （CnnPRL） was proposed to extract interpretable fine-grained emotional features progressively by using Convolutional Neural Network （CNN） based on acoustic features of speech. Firstly， interpretable shallow features were extracted manually and the optimal feature set was selected. Secondly， a cascaded CNN and dynamic fusion structure were proposed to refine the shallow features and learn deep emotional representations. Finally， a parallel heterogeneous CNN was constructed to extract complementary features at different scales to achieve multi-feature fusion by utilizing a fusion module， capture multi-granularity features， and integrate deep emotional information from different feature scales. Under the premise of ensuring time complexity， experimental results on the datasets IEMOCAP （Interactive EMOtional dyadic motion CAPture database）， CASIA （Institute of Automation， Chinese Academy of Sciences）， and EMODB （Berlin EMOtional DataBase） demonstrate that compared to the comparison methods such as SpeechFormer++， TLFMRF （Two-Layer Fuzzy Multiple Random Forest） and TIM-Net （Temporal-aware bI-direction Multi-scale Network）， CnnPRL achieves at least 0.86， 2.92， and 1.46 percentage points improvement in Weighted Average Recall （WAR） index， respectively， proving the effectiveness of CnnPRL. Ablation experimental results demonstrate that each module of CnnPRL contributes the overall performance improvement of the model.

Key words: Speech Emotion Recognition (SER), progressive emotional representation learning, Convolutional Neural Network (CNN), dynamic fusion, multi-scale fusion

摘要：

语音情感识别（SER）旨在赋予计算机准确识别语音信号中的情感状态的能力，而如何高效地表征语音中的情感特征一直是SER的研究热点。目前，大多数研究都致力于利用深度学习方法直接从原始语音或语谱图中学习最优特征，这种学习模式可以提取到更完整的特征信息，但忽略了对特定特征更深层细化信息的学习，同时不能保证特征的可解释性。为了解决上述问题，提出一种基于卷积神经网络的渐进式表征学习SER方法（CnnPRL），在语音声学特征的基础上利用卷积神经网络（CNN）渐进式地提取具有可解释性的精细化情感特征。首先，手工提取可解释的浅层特征并选择出最优的特征集；其次，提出级联CNN和动态融合结构，以细化浅层特征，并学习深层情感表征；最后，构建并行异构CNN提取不同尺度的互补特征，以利用融合模块实现多特征融合，捕获多粒度特征，并整合来自不同特征尺度的深层情感信息。实验结果表明，在保证时间复杂度的前提下，在数据集IEMOCAP （Interactive EMOtional dyadic motion CAPture database）、CASIA（Institute of Automation， Chinese Academy of Sciences）和EMODB（Berlin EMOtional DataBase）上，相较于SpeechFormer++、TLFMRF（Two-Layer Fuzzy Multiple Random Forest）和TIM-Net（Temporal-aware bI-direction Multi-scale Network）等对比方法，CnnPRL在指标加权平均召回率（WAR）上分别至少取得了0.86、2.92和1.46个百分点的提升，验证了CnnPRL的有效性；消融实验结果验证了CnnPRL的每个模块都有利于提升模型的整体性能。

关键词: 语音情感识别, 渐进式情感表征学习, 卷积神经网络, 动态融合, 多尺度融合

CLC Number:

TP183

Yonghong FAN, Heming HUANG. CnnPRL： progressive representation learning method for speech emotion recognition[J]. Journal of Computer Applications, 2025, 45(12): 3804-3812.

樊永红, 黄鹤鸣. 渐进式表征学习语音情感识别方法CnnPRL[J]. 《计算机应用》唯一官方网站, 2025, 45(12): 3804-3812.

Figures/Tables 13

References 32

[1]	陶建华，陈俊杰，李永伟. 语音情感识别综述［J］. 信号处理， 2023， 39（4）： 571-587.
	TAO J H， CHEN J J， LI Y W. Review on speech emotion recognition［J］. Journal of Signal Processing， 2023， 39（4）： 571-587.
[2]	BLASZKE M， KOSZEWSKI D. Determination of low-level audio descriptors of a musical instrument sound using neural network［C］// Proceedings of the 2020 Conference on Signal Processing： Algorithms， Architectures， Arrangements and Applications. Piscataway： IEEE， 2020： 138-141.
[3]	WANG X， DU P， CHEN D， et al. Change detection based on low-level to high-level features integration with limited samples［J］. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing， 2020， 13： 6260-6276.
[4]	HAN K， YU D， TASHEV I. Speech emotion recognition using deep neural network and extreme learning machine［C］// Proceedings of the INTERSPEECH 2014. ［S.l.］： International Speech Communication Association， 2014： 223-227.
[5]	ZHOU B， RICHARDSON K， NING Q， et al. Temporal reasoning on implicit events from distant supervision［C］// Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics： Human Language Technologies. Stroudsburg： ACL， 2021： 1361-1371.
[6]	ZHU Z， DAI W， HU Y， et al. Speech emotion recognition model based on Bi-GRU and focal loss ［J］. Pattern Recognition Letters， 2020， 140： 358-365.
[7]	RAJAMANI S T， RAJAMANI K T， MALLOL-RAGOLTA A， et al. A novel attention-based gated recurrent unit and its efficacy in speech emotion recognition［C］// Proceedings of the 2021 IEEE International Conference on Acoustics， Speech and Signal Processing. Piscataway： IEEE， 2021： 6294-6298.
[8]	LIU M， JOSEPH RAJ A N， RAJANGAM V， et al. Multiscale-multichannel feature extraction and classification through one-dimensional convolutional neural network for speech emotion recognition ［J］. Speech Communication， 2024， 156： No.103010.
[9]	MIAO X， LI Y， WEN M， et al. Fusing features of speech for depression classification based on higher-order spectral analysis［J］. Speech Communication， 2022， 143： 46-56.
[10]	GUO L， WANG L， DANG J， et al. Learning affective representations based on magnitude and dynamic relative phase information for speech emotion recognition［J］. Speech Communication， 2022， 136： 118-127.
[11]	ZHAO Z P， LI Q F， ZHANG Z X， et al. Combining a parallel 2D CNN with a self-attention dilated residual network for CTC-based discrete speech emotion recognition ［J］. Neural Networks， 2021， 141： 52-60.
[12]	LIU A T， LI S W， LEE H Y. TERA： self-supervised learning of Transformer encoder representation for speech［J］. IEEE/ACM Transactions on Audio， Speech， and Language Processing， 2021， 29： 2351-2366.
[13]	LIU Z， KANG X， REN F. Dual-TBNet： improving the robustness of speech features via dual-Transformer-BiLSTM for speech emotion recognition ［J］. IEEE/ACM Transactions on Audio， Speech， and Language Processing， 2023， 31： 2193-2203.
[14]	张钹，朱军，苏航.迈向第三代人工智能［J］.中国科学：信息科学，2020， 50（9）： 1281-1302.
	ZHANG B， ZHU J， SU H. Toward the third generation of artificial intelligence［J］. SCIENTIA SINICA Informationis， 2020， 50（9）： 1281-1302.
[15]	YE J X， WEN X C， WANG X Z， et al. GM-TCNet： gated multi-scale temporal convolutional network using emotion causality for speech emotion recognition［J］. Speech Communication， 2022， 145： 21-35.
[16]	ABADI M， BARHAM P， CHEN J， et al. TensorFlow： a system for large-scale machine learning［C］// Proceedings of the 12th USENIX Symposium on Operating Systems Design and Implementation. Berkeley： USENIX Association， 2016： 265-283.
[17]	KINGMA D P， BA J L. Adam： a method for stochastic optimization［EB/OL］. ［2024-11-03］..
[18]	CHEN W， XING X， XU X， et al. SpeechFormer++： a hierarchical efficient framework for paralinguistic speech processing［J］. IEEE/ACM Transactions on Audio， Speech， and Language Processing， 2023， 31： 775-788.
[19]	LIU Y， SUN H， GUAN W， et al. A discriminative feature representation method based on cascaded attention network with adversarial strategy for speech emotion recognition［J］. IEEE/ACM Transactions on Audio， Speech， and Language Processing， 2023， 31： 1063-1074.
[20]	THIRUMURU R， GURUGUBELLI K， VUPPALA A K. Novel feature representation using single frequency filtering and nonlinear energy operator for speech emotion recognition［J］. Digital Signal Processing， 2023， 120： No.103293.
[21]	CHANDOLA D， ALTARAWNEH E， JENKIN M， et al. SERC-GCN： speech emotion recognition in conversation using graph convolutional networks［C］// Proceedings of the 2024 IEEE International Conference on Acoustics， Speech and Signal Processing. Piscataway： IEEE， 2024： 76-80.
[22]	YE J， WEN X C， WEI Y， et al. Temporal modeling matters： a novel temporal emotional modeling approach for speech emotion recognition［C］// Proceedings of the 2023 IEEE International Conference on Acoustics， Speech and Signal Processing. Piscataway： IEEE， 2023： 1-5.
[23]	TAO H， GENG L， SHAN S， et al. Multi-stream convolution recurrent neural networks based on attention mechanism fusion for speech emotion recognition［J］. Entropy， 2022， 24（8）： No.1025.
[24]	HE J， REN L. Speech emotion recognition using XGBoost and CNN BLSTM with attention［C］// Proceedings of the 2021 IEEE SmartWorld， Ubiquitous Intelligence and Computing， Advanced and Trusted Computing， Scalable Computing and Communications， Internet of People and Smart City Innovation. Piscataway： IEEE， 2021： 154-159.
[25]	SUN L， FU S， WANG F. Decision tree SVM model with Fisher feature selection for speech emotion recognition［J］. EURASIP Journal on Audio， Speech， and Music Processing， 2019， 2019： No.2.
[26]	CHEN L， SU W， FENG Y， et al. Two-layer fuzzy multiple random forest for speech emotion recognition in human-robot interaction ［J］. Information Sciences， 2020， 509： 150-163.
[27]	LI S， XING X， FAN W， et al. Spatiotemporal and frequential cascaded attention networks for speech emotion recognition［J］. Neurocomputing， 2021， 448： 238-248.
[28]	CHEN Z， LI J， LIU H， et al. Learning multi-scale features for speech emotion recognition with connection attention mechanism［J］. Expert Systems with Applications， 2023， 214： No.118943.
[29]	FENG T， NARAYANAN S. Foundation model assisted automatic speech emotion recognition： transcribing， annotating， and augmenting［C］// Proceedings of the 2024 IEEE International Conference on Acoustics， Speech and Signal Processing. Piscataway： IEEE， 2024： 12116-12120.
[30]	MA Z， WU W， ZHENG Z， et al. Leveraging speech PTM， text LLM， and emotional TTS for speech emotion recognition［C］// Proceedings of the 2024 IEEE International Conference on Acoustics， Speech and Signal Processing. Piscataway： IEEE， 2024： 11146-11150.
[31]	ZHANG H， HUANG H， HAN H. MA-CapsNet-DA： speech emotion recognition based on MA-CapsNet using data augmentation［J］. Expert Systems with Applications， 2024， 244： No.122939.
[32]	CHEN W， XING X， XU X， et al. DST： deformable speech Transformer for emotion recognition［C］// Proceedings of the 2023 IEEE International Conference on Acoustics， Speech and Signal Processing. Piscataway： IEEE， 2024： 1-5.

SFR		MsDRL		SD	IEMOCAP		CASIA		EMODB
CasCNN	DF	PCN	MsF	SD	WAR	UAR	WAR	UAR	WAR	UAR
√	√			√	69.03	67.70	84.17	84.17	85.98	84.17
		√	√	√	68.83	67.95	85.00	85.00	86.92	85.94
√		√		√	69.13	67.83	84.58	84.58	87.85	87.88
√	√	√		√	70.04	69.06	85.83	85.83	89.72	88.33
√		√	√	√	70.14	69.62	86.67	86.67	88.79	87.73
√	√	√	√		70.65	70.39	87.50	87.50	87.85	86.37
√	√	√	√	√	71.36	70.59	88.75	88.75	90.65	89.64

SFR		MsDRL		SD	IEMOCAP		CASIA		EMODB
CasCNN	DF	PCN	MsF	SD	WAR	UAR	WAR	UAR	WAR	UAR
√	√			√	69.03	67.70	84.17	84.17	85.98	84.17
		√	√	√	68.83	67.95	85.00	85.00	86.92	85.94
√		√		√	69.13	67.83	84.58	84.58	87.85	87.88
√	√	√		√	70.04	69.06	85.83	85.83	89.72	88.33
√		√	√	√	70.14	69.62	86.67	86.67	88.79	87.73
√	√	√	√		70.65	70.39	87.50	87.50	87.85	86.37
√	√	√	√	√	71.36	70.59	88.75	88.75	90.65	89.64

数据集	方法	模型结构	WAR	UAR
IEMOCAP	MCMA^［10］	CNN	57.58	58.07
	Dual-TBNet^［13］	Transformer+BiLSTM	64.80	64.00
	SFF-NEC^［20］	SVM	65.78	64.27
	SERC-GCN^［21］	Graph Convolution	66.85	—
	Foundation model^［29］	Transformer-based	67.45	—
	PTM-based SER^［30］	Speech Pre-trained Model	68.85	—
	TIM-Net^［22］	TCN	69.00	68.29
	BiGRU-Focal^［6］	CNN+BiGRU	69.73	68.75
	SpeechFormer++^［18］	Transformer-based	70.50	—
	DST^［32］	Transformer-based	71.80	—
	CnnPRL	CNN	71.36	70.59
CASIA	MSCRNN-A^［23］	CNN+BiLSTM	60.75	60.75
	AS-CANet^［19］	CNN	67.08	67.08
	CNN-LSTM^［24］	CNN+LSTM	74.17	74.17
	MA-CapsNet-DA^［31］	Capsule Network	76.28	—
	DT-SVM^［25］	SVM	85.08	85.08
	TLFMRF^［26］	Random Forest	85.83	85.83
	CnnPRL	CNN	88.75	88.75
EMODB	MA-CapsNet-DA^［31］	Capsule Network	77.89	—
	SFF-NEC^［20］	SVM	82.84	81.45
	ST-FCANet^［27］	CNN	83.30	82.10
	Dual-TBNet^［13］	Transformer+BiLSTM	84.10	—
	DT-SVM^［25］	SVM	87.55	—
	AMSNet^［28］	CNN+BiLSTM	88.34	88.56
	MSCRNN-A^［23］	CNN+BiLSTM	88.41	87.95
	TIM-Net^［22］	TCN	89.19	—
	CnnPRL	CNN	90.65	89.64
	MCDNet^［8］	CNN	96.25	—
	CnnPRL^*	CNN	99.07	99.27

数据集	方法	模型结构	WAR	UAR
IEMOCAP	MCMA^［10］	CNN	57.58	58.07
	Dual-TBNet^［13］	Transformer+BiLSTM	64.80	64.00
	SFF-NEC^［20］	SVM	65.78	64.27
	SERC-GCN^［21］	Graph Convolution	66.85	—
	Foundation model^［29］	Transformer-based	67.45	—
	PTM-based SER^［30］	Speech Pre-trained Model	68.85	—
	TIM-Net^［22］	TCN	69.00	68.29
	BiGRU-Focal^［6］	CNN+BiGRU	69.73	68.75
	SpeechFormer++^［18］	Transformer-based	70.50	—
	DST^［32］	Transformer-based	71.80	—
	CnnPRL	CNN	71.36	70.59
CASIA	MSCRNN-A^［23］	CNN+BiLSTM	60.75	60.75
	AS-CANet^［19］	CNN	67.08	67.08
	CNN-LSTM^［24］	CNN+LSTM	74.17	74.17
	MA-CapsNet-DA^［31］	Capsule Network	76.28	—
	DT-SVM^［25］	SVM	85.08	85.08
	TLFMRF^［26］	Random Forest	85.83	85.83
	CnnPRL	CNN	88.75	88.75
EMODB	MA-CapsNet-DA^［31］	Capsule Network	77.89	—
	SFF-NEC^［20］	SVM	82.84	81.45
	ST-FCANet^［27］	CNN	83.30	82.10
	Dual-TBNet^［13］	Transformer+BiLSTM	84.10	—
	DT-SVM^［25］	SVM	87.55	—
	AMSNet^［28］	CNN+BiLSTM	88.34	88.56
	MSCRNN-A^［23］	CNN+BiLSTM	88.41	87.95
	TIM-Net^［22］	TCN	89.19	—
	CnnPRL	CNN	90.65	89.64
	MCDNet^［8］	CNN	96.25	—
	CnnPRL^*	CNN	99.07	99.27

子模块	CasCNN	DF	PCN	MsF	FLOPs/10⁶
模块消融	√				0.29
	√	√			0.34
	√	√	√		1.08
CnnPRL	√	√	√	√	1.31

CnnPRL： progressive representation learning method for speech emotion recognition

渐进式表征学习语音情感识别方法CnnPRL

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 13

References 32

Related Articles 15

Recommended Articles

Metrics

数据集	方法	愤怒	快乐	兴奋	中性	悲伤	无聊	厌恶	恐惧	惊讶
IEMOCAP	MCMA^［10］	65.10	—	53.18	57.85	56.18	—	—	—	—
	Dual-TBNet^［13］	62.00	—	59.00	71.00	64.00	—	—	—	—
	BiGRU-Focal^［6］	66.54	—	55.31	71.85	60.23	—	—	—	—
	CnnPRL	67.87	—	61.54	77.49	74.65	—	—	—	—
CASIA	MSCRNN-A^［23］	74.00	59.00	—	71.00	76.50	—	—	38.50	45.50
	DT-SVM^［25］	90.00	88.50	—	92.50	78.00	—	—	74.50	87.00
	TLFMRF^［26］	93.00	77.00	—	94.00	88.00	—	—	67.00	98.00
	CnnPRL	95.00	90.00	—	95.00	87.50	—	—	70.00	95.00
EMODB	Dual-TBNet^［13］	75.00	82.00	—	75.00	100.00	88.00	100.00	90.00	—
	DT-SVM^［25］	91.80	85.00	—	94.50	83.00	89.85	80.45	88.30	—
	AMSNet^［28］	92.00	91.00	—	85.00	92.00	85.00	—	90.00	—
	MSCRNN-A^［23］	97.00	72.87	—	87.55	91.35	92.57	86.87	81.29	—
	CnnPRL	96.15	85.71	—	87.50	100.00	87.50	77.78	92.86	—
	MCDNet^［8］	97.37	94.74	—	100.00	100.00	100.00	92.86	95.24	—
	CnnPRL^*	98.04	100.00	—	96.88	100.00	100.00	100.00	100.00	—

[1]	Chao SHI, Yuxin ZHOU, Qian FU, Wanyu TANG, Ling HE, Yuanyuan LI. Action recognition algorithm for ADHD patients using skeleton and 3D heatmap [J]. Journal of Computer Applications, 2025, 45(9): 3036-3044.
[2]	Hongjun ZHANG, Gaojun PAN, Hao YE, Yubin LU, Yiheng MIAO. Multi-source heterogeneous data analysis method combining deep learning and tensor decomposition [J]. Journal of Computer Applications, 2025, 45(9): 2838-2847.
[3]	Peng PENG, Ziting CAI, Wenling LIU, Caihua CHEN, Wei ZENG, Baolai HUANG. Speech emotion recognition method based on hybrid Siamese network with CNN and bidirectional GRU [J]. Journal of Computer Applications, 2025, 45(8): 2515-2521.
[4]	Jinhao LIN, Chuan LUO, Tianrui LI, Hongmei CHEN. Thoracic disease classification method based on cross-scale attention network [J]. Journal of Computer Applications, 2025, 45(8): 2712-2719.
[5]	Yongpeng TAO, Shiqi BAI, Zhengwen ZHOU. Neural architecture search for multi-tissue segmentation using convolutional and transformer-based networks in glioma segmentation [J]. Journal of Computer Applications, 2025, 45(7): 2378-2386.
[6]	Yingjun ZHANG, Weiwei YAN, Binhong XIE, Rui ZHANG, Wangdong LU. Gradient-discriminative and feature norm-driven open-world object detection [J]. Journal of Computer Applications, 2025, 45(7): 2203-2210.
[7]	Yingtao CHEN, Kangkang FANG, Jin’ao ZHANG, Haoran LIANG, Huanbin GUO, Zhaowen QIU. Segmentation network of coronary artery structure from CT angiography images based on multi-scale spatial features [J]. Journal of Computer Applications, 2025, 45(6): 2007-2015.
[8]	Junyan ZHANG, Yiming ZHAO, Bing LIN, Yunping WU. Chinese image captioning method based on multi-level visual and dynamic text-image interaction [J]. Journal of Computer Applications, 2025, 45(5): 1520-1527.
[9]	Dan WANG, Wenhao ZHANG, Lijuan PENG. Channel estimation of reconfigurable intelligent surface assisted communication system based on deep learning [J]. Journal of Computer Applications, 2025, 45(5): 1613-1618.
[10]	Baohua YUAN, Jialu CHEN, Huan WANG. Medical image segmentation network integrating multi-scale semantics and parallel double-branch [J]. Journal of Computer Applications, 2025, 45(3): 988-995.
[11]	Dixin WANG, Jiahao WANG, Min LI, Hao CHEN, Guangyao HU, Yu GONG. Abnormal attack detection for underwater acoustic communication network [J]. Journal of Computer Applications, 2025, 45(2): 526-533.
[12]	Jianhua REN, Jiahui CAO, Di JIA. Hand pose estimation based on mask prompts and attention [J]. Journal of Computer Applications, 2025, 45(12): 4012-4020.
[13]	Xinran XU, Shaobing ZHANG, Miao CHENG, Yang ZHANG, Shang ZENG. Bearings fault diagnosis method based on multi-pathed hierarchical mixture-of-experts model [J]. Journal of Computer Applications, 2025, 45(1): 59-68.
[14]	Yun LI, Fuyou WANG, Peiguang JING, Su WANG, Ao XIAO. Uncertainty-based frame associated short video event detection method [J]. Journal of Computer Applications, 2024, 44(9): 2903-2910.
[15]	Hong CHEN, Bing QI, Haibo JIN, Cong WU, Li’ang ZHANG. Class-imbalanced traffic abnormal detection based on 1D-CNN and BiGRU [J]. Journal of Computer Applications, 2024, 44(8): 2493-2499.