Parameters Of Convolutional Neural Network Research Articles

Real-Time CNN Training and Compression for Neural-Enhanced Adaptive Live Streaming.

We propose a real-time convolutional neural network (CNN) training and compression method for delivering high-quality live video even in a poor network environment. The server delivers a low-resolution video segment along with the corresponding CNN for super resolution (SR), after which the client applies the CNN to the segment in order to recover high-resolution video frames. To generate a trained CNN corresponding to a video segment in real-time, our method rapidly increases the training accuracy by promoting the overfitting property of the CNN while also using curriculum-based training. In addition, assuming that the pretrained CNN is already downloaded on the client side, we transfer only residual values between the updated and pretrained CNN parameters. These values can be quantized with low bits in real time while minimizing the amount of loss, as the distribution range is significantly narrower than that of the updated CNN. Quantitatively, our neural-enhanced adaptive live streaming pipeline (NEALS) achieves higher SR accuracy and a lower CNN compression loss rate within a constrained training time compared to the state-of-the-art CNN training and compression method. NEALS achieves 15 to 48% higher quality of the user experience compared to state-of-the-art neural-enhanced live streaming systems.

An attribution graph-based interpretable method for CNNs

Convolutional Neural Networks (CNNs) have demonstrated outstanding performance in various domains, such as face recognition, object detection, and image segmentation. However, the lack of transparency and limited interpretability inherent in CNNs pose challenges in fields such as medical diagnosis, autonomous driving, finance, and military applications. Several studies have explored the interpretability of CNNs and proposed various post-hoc interpretable methods. The majority of these methods are feature-based, focusing on the influence of input variables on outputs. Few methods undertake the analysis of parameters in CNNs and their overall structure. To explore the structure of CNNs and intuitively comprehend the role of their internal parameters, we propose an Attribution Graph-based Interpretable method for CNNs (AGIC) which models the overall structure of CNNs as graphs and provides interpretability from global and local perspectives. The runtime parameters of CNNs and feature maps of each image sample are applied to construct attribution graphs (At-GCs), where the convolutional kernels are represented as nodes and the SHAP values between kernel outputs are assigned as edges. These At-GCs are then employed to pretrain a newly designed heterogeneous graph encoder based on Deep Graph Infomax (DGI). To comprehensively delve into the overall structure of CNNs, the pretrained encoder is used for two types of interpretable tasks: (1) a classifier is attached to the pretrained encoder for the classification of At-GCs, revealing the dependency of At-GC’s topological characteristics on the image sample categories, and (2) a scoring aggregation (SA) network is constructed to assess the importance of each node in At-GCs, thus reflecting the relative importance of kernels in CNNs. The experimental results indicate that the topological characteristics of At-GC exhibit a dependency on the sample category used in its construction, which reveals that kernels in CNNs show distinct combined activation patterns for processing different image categories, meanwhile, the kernels that receive high scores from SA network are crucial for feature extraction, whereas low-scoring kernels can be pruned without affecting model performance, thereby enhancing the interpretability of CNNs.

A new multi-objective hyperparameter optimization algorithm for COVID-19 detection from x-ray images

The coronavirus occurred in Wuhan (China) first and it was declared a global pandemic. To detect coronavirus X-ray images can be used. Convolutional neural networks (CNNs) are used commonly to detect illness from images. There can be lots of different alternative deep CNN models or architectures. To find the best architecture, hyper-parameter optimization can be used. In this study, the problem is modeled as a multi-objective optimization (MOO) problem. Objective functions are multi-class cross entropy, error ratio, and complexity of the CNN network. For the best solutions to the objective functions, multi-objective hyper-parameter optimization is made by NSGA-III, NSGA-II, R-NSGA-II, SMS-EMOA, MOEA/D, and proposed Swarm Genetic Algorithms (SGA). SGA is a swarm-based algorithm with a cross-over process. All six algorithms are run and give Pareto optimal solution sets. When the figures obtained from the algorithms are analyzed and algorithm hypervolume values are compared, SGA outperforms the NSGA-III, NSGA-II, R-NSGA-II, SMS-EMOA, and MOEA/D algorithms. It can be concluded that SGA is better than others for multi-objective hyper-parameter optimization algorithms for COVID-19 detection from X-ray images. Also, a sensitivity analysis has been made to understand the effect of the number of the parameters of CNN on model success.

Prediction of photovoltaic power generation based on a hybrid model

In order to fully exploit the relationship between temporal features in photovoltaic power generation data and improve the prediction accuracy of photovoltaic power generation, a photovoltaic power generation forecasting method is proposed based on a hybrid model of the convolutional neural network (CNN) and extreme gradient boost (XGBoost). Taking the historical data of China’s photovoltaic power plants as a sample, the high-dimensional mapping relationship of photovoltaic power generation variables is extracted based on the convolutional layer and pooling layer of the CNN network to construct a high-dimensional time-series feature vector, which is an input for the XGBoost. A photovoltaic power generation prediction model is established based on CNN-XGBoost by training CNN and XGBoost parameters. Since it is difficult for a single model to achieve optimal prediction accuracy under different weather conditions, the k-means clustering algorithm is used to group the power datasets and train independent models to improve prediction accuracy. Through the actual data verification of photovoltaic power plants, the proposed photovoltaic power generation prediction model can accurately predict the power, which shows high prediction accuracy and generalization ability compared with other methods.

Disease detection and classification in pomegranate fruit using hybrid convolutional neural network with honey badger optimization algorithm

ABSTRACT Fruit cultivation plays a pivotal role in improvement of the agricultural economy. Pomegranate is a nutritionally rich fruit that is highly valuable because of its excellent antioxidant properties, richness in vitamins and fiber. Pomegranate is affected by diseases, which can severely affect its yield and quality. Timely disease detection is vital for effective disease control in pomegranate orchards. This study presents a novel approach for disease detection in pomegranate and categorization of pomegranate using a hybrid Convolutional Neural Network (CNN) coupled with the Honey Badger Optimization Algorithm (HBOA). The proposed hybrid CNN-HBOA approach leverages the strengths of both CNN and metaheuristic optimization algorithm HBOA. The CNN model was designed to extract high-level features from images of pomegranate affected by various diseases, and HBOA was employed to fine-tune the CNN parameters to improve the accuracy and robustness of the model. Initially, the quality of the images was enhanced using a contrast enhancement technique, and the images were segmented using the Detectron2 algorithm. Finally, the extracted features were fed as input to the disease detection stage wherein the CNN classifies the disease. Furthermore, the classification accuracy of the CNN model was enhanced using the HBOA algorithm. The performance of the hybrid CNN-HBOA algorithm was evaluated using a vast collection of images depicting multiple disease categories of pomegranate. Experimental results demonstrate the superiority of the hybrid CNN-HBOA system, which achieves a detection accuracy of 99.58%, precision of 100%, recall of 99.71% and F1 score of 99.75% compared to other existing state-of-the-art models.

An optimized convolutional neural network architecture for lung cancer detection.

Lung cancer, the treacherous malignancy affecting the respiratory system of a human body, has a devastating impact on the health and well-being of an individual. Due to the lack of automated and noninvasive diagnostic tools, healthcare professionals look forward toward biopsy as a gold standard for diagnosis. However, biopsy could be traumatizing and expensive process. Additionally, the limited availability of dataset and inaccuracy in diagnosis is a major drawback experienced by researchers. The objective of the proposed research is to develop an automated diagnostic tool for screening of lung cancer using optimized hyperparameters such that convolutional neural network (CNN) model generalizes well for universally obtained computerized tomography (CT) slices of lung pathologies. The aforementioned objective is achieved in the following ways: (i) Initially, a preprocessing methodology specific to lung CT scans is formulated to avoid the loss of information due to random image smoothing, and (ii) a sine cosine algorithm optimization algorithm (SCA) is integrated in the CNN model, to optimally select the tuning parameters of CNN. The error rate is used as an objective function, and the SCA algorithm tries to minimize. The proposed method successfully achieved an average classification accuracy of 99% in classification of lung scans in normal, benign, and malignant classes. Further, the generalization ability of the proposed model is tested on unseen dataset, thereby achieving promising results. The quantitative results prove the efficacy of the system to be used by radiologists in a clinical scenario.

Enhancing prediction accuracy of Remaining Useful Life in lithium-ion batteries: A deep learning approach with Bat optimizer

Remaining Useful Life (RUL) prediction in lithium-ion batteries is crucial for assessing battery performance. Despite the popularity of deep learning methods for RUL prediction, their complex architectures often pose challenges in interpretation and resource consumption. We propose a novel approach that combines the interpretability of a convolutional neural network (CNN) with the efficiency of a bat-based optimizer. CNN extracts battery data features and characterizes degradation kinetics, while the optimizer refines CNN parameters. Tested on NASA PCoE data, our method achieves exceptional results with minimal computational burden and fewer parameters. It outperforms traditional approaches, yielding an R2-score of 0.9987120, an MAE of 0.004397067 Ah, and a low RMSE of 0.00656 Ah. The proposed model outperforms traditional deep learning models, as confirmed by comparative analysis.

Identification method of internal leakage in nuclear power plants valves using convolutional block attention module

Valves are indispensable fluid control devices widely employed in Nuclear Power Plant (NPP) structures. Due to prolonged exposure to high-temperature environments, the internal sealing components become susceptible to thermal deformation or wear, leading to compromised sealing and potential leakage accidents in the valves. Therefore, detecting internal leakage in valves holds great significance. We propose an effective acoustic emission detection device designed for identifying internal leakage in valves within high-temperature environments. A valve internal leakage monitoring system based on the Acoustic Emission (AE) method is developed, utilizing high-temperature piezoelectric transducers as AE sensors. To efficiently and rapidly identify leakage states, an artificial intelligence (AI) algorithm incorporating a Convolutional Neural Network (CNN) with the Convolutional Block Attention Module (CBAM) is employed. The approach integrates both a channel attention module and a spatial attention module to capture global and local features, respectively. Normalized spectral data is used as the training set to optimize the CNN parameters. The performance of this method is then compared with BAM and SENET. The results indicate that CBAM can effectively enhance and suppress features by allocating weights among channels according to task requirements. Moreover, it can more efficiently identify multiple damage features under the constraints of limited computing resources.

Rapid quantitative analysis of cotton-polyester blended fabrics using near-infrared spectroscopy combined with CNN-LSTM

Accurate and rapid quantitative analysis of fiber content is important for blended fabric quality control and supervision. In this paper, a deep learning method based on an improved hybrid network of convolutional neural network (CNN) and long short-term memory (LSTM) to fast and accurate determine cotton content of cotton-polyester blended fabrics using near-infrared (NIR) spectroscopy. The cotton-polyester blended fabric samples were partitioned into training set, validation set and testing set in the ratio of 6:2:2 to develop the quantitative model of cotton content of blended fabric. Firstly, a one-dimensional CNN architecture is designed and modeled to obtain the CNN parameters and to evaluate the spectral preprocessing methods. Then, the CNN structure is modified and improved using LSTM to obtain a CNN-LSTM structure. The root mean square error (RMSE) of the validation set obtained by CNN model is 0.83%, which is 34.03% lower than that of the traditional partial least squares (PLS) model, and the RMSE of the testing set obtained by CNN model is 0.88%, which is 25.88% lower than that of the PLS model. The RMSE of the validation set obtained by CNN-LSTM model is 0.65%, which is 47.85% lower than that of the PLS model, and the RMSE of the testing set obtained by CNN-LSTM model is 0.75%, which is 36.57% lower than that of the PLS model. Compared with PLS and CNN models, the CNN-LSTM model can more accurately analyze the cotton content proportion of cotton-polyester blended fabrics, which provides a new fast and non-invasive method for the quantitative detection of fabrics.

Performance analysis of genetically optimized 1D-convolutional neural network architecture for rotor system fault detection and diagnosis

Rotor mass imbalance is a critical problem in industries, which can lead to a breakdown or a catastrophic failure if unattended. Hence, Industry 4.0 uses deep learning approaches like convolutional neural network (CNN) for rotor fault detection and diagnosis of Industrial machines and turbomachinery where multifunctional structures are used. In machinery fault diagnosis, manual tuning of CNN hyperparameters is widely used, which is a tedious, time-consuming, and challenging task to achieve effective feature extraction. Additionally, overfitting of CNN during its training is an obstacle to achieving higher prediction accuracy. Hence, to address these issues, this research focuses on multi-class mass imbalance fault diagnosis using genetically optimized deep learning architecture of 1D-CNN to achieve higher diagnosis capability. The performance of the proposed methodology is evaluated through various case studies using experimental data from the in-house developed test rig. The test results are compared against various 1D-CNN architectures in which the hyperparameters and effective dropout layer positioning are genetically optimized. Also, the performance is benchmarked against standard machine learning algorithms. The results show that genetic algorithm (GA)-optimized 1D-CNN with dropout achieves highest fault prediction accuracy of 97.47% with reduced depth of CNN (with three convolution layers) compared to 84.16% of manually tunned CNN architecture (with seven convolutional layers) and outperformed standard machine learning algorithms. Best drop positioning (end of feature extraction part) reduced the learnable parameters to 95.9% with the highest prediction accuracy of 97.47% compared to adding dropout at all the CNN layers. The proposed GA-optimized 1D-CNN with effective dropout positioning eliminates human intervention and reduces the depth of CNN architecture; this, in turn, reduces computational load and time by reducing learnable parameters of CNN (network weights) with the highest prediction accuracy. Hence, the proposed approach shows promise in enhancing the performance and contributing to the advancement of 1D-CNN for rotor system fault detection and diagnosis.

A Novel Fault Diagnosis Method for a Power Transformer Based on Multi-Scale Approximate Entropy and Optimized Convolutional Networks.

Dissolved gas analysis (DGA) in transformer oil, which analyzes its gas content, is valuable for promptly detecting potential faults in oil-immersed transformers. Given the limitations of traditional transformer fault diagnostic methods, such as insufficient gas characteristic components and a high misjudgment rate for transformer faults, this study proposes a transformer fault diagnosis model based on multi-scale approximate entropy and optimized convolutional neural networks (CNNs). This study introduces an improved sparrow search algorithm (ISSA) for optimizing CNN parameters, establishing the ISSA-CNN transformer fault diagnosis model. The dissolved gas components in the transformer oil are analyzed, and the multi-scale approximate entropy of the gas content under different fault modes is calculated. The computed entropy values are then used as feature parameters for the ISSA-CNN model to derive diagnostic results. Experimental data analysis demonstrates that multi-scale approximate entropy effectively characterizes the dissolved gas components in the transformer oil, significantly improving the diagnostic efficiency. Comparative analysis with BPNN, ELM, and CNNs validates the effectiveness and superiority of the proposed ISSA-CNN diagnostic model across various evaluation metrics.

Design and Implementation of Novel Hybrid and Multiscale- Assisted CNN and ResNet Using Heuristic Advancement of Adaptive Deep Segmentation for Iris Recognition

Due to its significant applications in security, the iris recognition process has been considered as the most active research area over the last few decades. In general, the iris recognition framework has been crucially utilized for various security applications because it includes a set of features as well as does not alter its character according to the time. In recent times, emerging deep learning techniques have attained huge success, particularly in the field of the iris recognition framework model. Moreover, in considering the field of iris recognition, there is no possibility for the remarkable capability of the deep learning model as well as to attain superior performance. To handle the issues in the conventional model of iris recognition, a novel heuristic-aided deep learning framework has been implemented for recognizing the iris system. Initially, the required source iris images are gathered from the data sources. It is then followed by the pre-processing stage, where the pre-processed image is obtained. Consequently, the image segmentation process is carried out by Adaptive Deeplabv3+layers, in which the parameters are optimized using the Modified Weighted Flow Direction Algorithm (MWFDA). Finally, the iris recognition is accomplished by hybrid Hybridization of Multiscale Dilated-Assisted Learning (MDAL) that will be composed of both a Convolutional Neural Network (CNN) and a Residual Network (ResNet). To achieve optimal recognition results, the parameters in CNN and ResNet are tuned optimally by using MWFDA. The experimental results are estimated with the help of distinct measures. Contrary to conventional methods, the empirical results prove that the recommended model achieves the desired value to enhance the recognition performance.

Mobile detection of cataracts with an optimised lightweight deep Edge Intelligent technique

AbstractTesting the visual field is a valuable diagnostic tool for identifying eye conditions such as cataract, glaucoma, and retinal disease. Its quick and straightforward testing process has become an essential component in our efforts to prevent blindness. Still, the device must be accessible to the general masses. This research has developed a machine learning model that can work with Edge devices like smartphones. As a result, it is opening the opportunity to integrate the disease‐detecting model into multiple Edge devices to automate their operation. The authors intend to use convolutional neural network (CNN) and deep learning to deduce which optimisers have the best results when detecting cataracts from live photos of eyes. This is done by comparing different models and optimisers. Using these methods, a reliable model can be obtained that detects cataracts. The proposed TensorFlow Lite model constructed by combining CNN layers and Adam in this study is called Optimised Light Weight Sequential Deep Learning Model (SDLM). SDLM is trained using a smaller number of CNN layers and parameters, which gives SDLM its compatibility, fast execution time, and low memory requirements. The proposed Android app, I‐Scan, uses SDLM in the form of TensorFlow Lite for demonstration of the model in Edge devices.

Integrating Convolution and Sparse Coding for Learning Low-Dimensional Discriminative Image Representations.

This work investigates the problem of efficiently learning discriminative low-dimensional (LD) representations of multiclass image objects. We propose a generic end-to-end approach that jointly optimizes sparse dictionary and convolutions for learning LOW-dimensional discriminative image representations, named SparConvLow, taking advantage of convolutional neural networks (CNNs), dictionary learning, and orthogonal projections. The whole learning process can be summarized as follows. First, a CNN module is employed to extract high-dimensional (HD) preliminary convolutional features. Second, to avoid the high computational cost of direct sparse coding on HD CNN features, we learn sparse representation (SR) over a task-driven dictionary in the space with the feature being orthogonally projected. We then exploit the discriminative projection on SR. The whole learning process is consistently treated as an end-to-end joint optimization problem of trace quotient maximization. The cost function is well-defined on the product of the CNN parameters space, the Stiefel manifold, the Oblique manifold, and the Grassmann manifold. By using the explicit gradient delivery, the cost function is optimized via a geometrical stochastic gradient descent (SGD) algorithm along with the chain rule and the backpropagation. The experimental results show that the proposed method can achieve a highly competitive performance with the state-of-the-art (SOTA) image classification, object categorization, and face recognition methods, under both supervised and semi-supervised settings. The code is available at https://github.com/MVPR-Group/SparConvLow.

Convolutional neural networks applied to dissolved gas analysis for power transformers condition monitoring

In this contribution a methodology to diagnose transformer faults based on Dissolved Gas Analysis (DGA) by using a convolutional neural network (CNN) is proposed. The algorithm to transform the gas contents (resulting from the DGA analysis) into feature maps is introduced, and the resulting feature maps are the input of the CNN. In order to take into account the fact that the data set is imbalanced, the improved Synthetic Minority Over-Sampling Technique (SMOTE) is combined with the data cleaning technique to protect the CNN from training bias. The effect of the CNN architecture on the classification performance is also investigated to determine the optimal CNN parameters. All the above mentioned possibilities are tested and their performance investigated; in addition, a final test on the IEC TC 10 transformer fault database validates the accuracy and the generalization potential of the proposed methodology.

An Indoor Fire Detection Method Based on Multi-Sensor Fusion and a Lightweight Convolutional Neural Network.

Indoor fires pose significant threats in terms of casualties and economic losses globally. Thus, it is vital to accurately detect indoor fires at an early stage. To improve the accuracy of indoor fire detection for the resource-constrained embedded platform, an indoor fire detection method based on multi-sensor fusion and a lightweight convolutional neural network (CNN) is proposed. Firstly, the Savitzky-Golay (SG) filter is used to clean the three types of heterogeneous sensor data, then the cleaned sensor data are transformed by means of the Gramian Angular Field (GAF) method into matrices, which are finally integrated into a three-dimensional matrix. This preprocessing stage will preserve temporal dependency and enlarge the characteristics of time-series data. Therefore, we could reduce the number of blocks, channels and layers in the network, leading to a lightweight CNN for indoor fire detection. Furthermore, we use the Fire Dynamic Simulator (FDS) to simulate data for the training stage, enhancing the robustness of the network. The fire detection performance of the proposed method is verified through an experiment. It was found that the proposed method achieved an impressive accuracy of 99.1%, while the number of CNN parameters and the amount of computation is still small, which is more suitable for the resource-constrained embedded platform of an indoor fire detection system.

Identification of Organic and Non-Organic Waste with Computer Image Recognition using Convolutionalneural Network with Efficient-Net-B0 Architecture

This study aims to develop a method for identifying organic and non-organic waste using a computer image recognition technique based on Convolutional Neural Network (CNN) with Efficient-Net-B0 architecture. Efficient and accurate waste identification is important in sustainable waste management. The primary goal of this research is to distinguish between organic and non-organic waste in images. Manually labeling waste images as organic or non-organic can be a time-consuming and error-prone task. Configuring and fine-tuning the EfficientNet-B0 architecture and CNN parameters for optimal performance can be a complex and iterative process. Hyperparameter tuning may be needed. Ensuring accurate labels is essential for training a reliable model. The choice of using the Convolutional Neural Network (CNN) with the EfficientNet-B0 architecture is a crucial part of the solution. EfficientNet-B0 is known for its balance between accuracy and computational efficiency. The use of CNNs and EfficientNet-B0 for this task indicates the system's ability to discern visual differences between the two waste types. The method proposed in this study utilizes CNN's ability to study important features of waste images to recognize various types of waste. This research includes the waste data collection stage which includes organic and non-organic waste in the form of 2D images. To evaluate the performance of the proposed method, a test was carried out using a waste dataset taken from a predetermined environment. The test results show that the proposed method is able to identify organic and non-organic waste with a high degree of accuracy. In test scenarios, this method achieves an accuracy of 98%, which demonstrates its ability to effectively identify the type of waste. Through the use of CNN-based computer image recognition techniques with the Efficient-Net-B0 architecture, this research succeeded in solving the problem of identifying organic and non-organic waste automatically and accurately. The proposed method has the potential to be applied in more efficient waste management systems, helps minimize human identification errors, and makes a positive contribution to environmental protection efforts. This research is expected to be the basis for further development in the introduction and management of waste in a sustainable manner.

Intelligent damage diagnosis method for offshore platforms based on enhanced stabilization diagrams and convolutional neural network

Considering that it is difficult to evaluate the damage state of offshore platform structures under environmental excitation by stochastic subspace identification (SSI) stability diagrams alone, an intelligent damage diagnosis method based on enhanced stability diagrams and convolutional neural network (CNN) is proposed. The data-driven SSI algorithm and covariance-driven SSI algorithm are utilized to identify stability diagrams of monitoring data, and the stability diagrams of the two algorithms are superimposed together for image enhancement. Further, the enhanced stability diagrams are used as input samples for CNN training to distinguish the damage state of the structure. In the meanwhile, the whale optimization algorithm is employed to optimize the hyper parameters of CNN to ulteriorly improve the recognition performance. The final test accuracy of CNN is 97.20%, and is 13.09% higher than before hyper parameter optimization, which indicates that the damage diagnosis method based on enhanced stability diagrams and CNN is reasonable and effective, and is expected to be applied to real-time damage diagnosis of offshore platform structures.

Crop Yield Prediction Using With The Simulated Annealing Convolutional Neural Networks (SACNNS) Optimization Algorithm

This paper introduces an innovative approach for precise crop yield prediction in Tamil Nadu, India, a region with diverse agricultural conditions. The Simulated Annealing Convolutional Neural Networks (SACNNs) optimization algorithm is employed to enhance the performance of Convolutional Neural Networks (CNNs). By leveraging a comprehensive agricultural dataset that includes weather information, soil characteristics, and historical crop yields, SACNNs fine-tunes CNN architecture and parameters for improved prediction accuracy. The model excels at predicting crop yields at a local level, accommodating the unique agricultural conditions of different districts in Tamil Nadu. Experimental results highlight SACNNs' superiority in prediction accuracy and robustness compared to conventional CNNs and other optimization methods. These findings offer valuable insights for informed decision-making in agricultural planning and resource allocation, benefiting farmers, policymakers, and stakeholders in the region.

Parameter optimization in convolutional neural network for improving performance of facial expression recognition

The recognition of similar facial expressions presents a notable challenge, necessitating a focus on the parameters within the fundamental Convolutional Neural Network (CNN) architecture, which serves as a cornerstone in the field of image classification. This research endeavor aims to enhance the model's capacity for facial expression recognition by employing a controlled variable method to examine two specific parameters in a self-designed small CNN: the number of filters and convolutional layers. More specifically, while the filters were fixed at 3, the layers varied from 3 to 6 to 9. Similarly, as the number of the filters rose to 6, the number of the layers also incremented from 3 to 6 to 9. Furthermore, while the number of the filters reached 12, the number of the layers went from 3 to 6 to 9 too. Finally, with the filters increasing to 24, the layers rose from 3 to 6 to 9 as well. Experimental results indicate that both increasing the number of filters and convolutional layers can increment the performance of model in facial expression recognition. Furthermore, increasing the number of filters can exert a more prominent influence on improving the accuracy of facial expression recognition.