A high-performance convolution block oriented accelerator for MBConv-Based CNNs

Shang, Jiangwei; Zhang, Kun; Zhang, Zhan; Li, Chuanyou; Liu, Hongwei

doi:10.1016/j.vlsi.2022.10.012

Cited by 4 publications

(3 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The MBConv block was first introduced from MobileNetV2 [55], and it is widely used in lightweight CNN designs because the number of parameters and computational effort of the MBConv block is less than the regular convolution [56], and the overall architecture of the MBConv block is shown in figure 3. As can be seen in figure 3, the MBConv block first upscales the input features by 1 × 1 convolution to obtain a more complex feature representation.…”

Section: Mbconv Blockmentioning

confidence: 99%

A novel multi-sensor hybrid fusion framework

Du,

Wang,

Zhang

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

Multi-sensor data fusion has emerged as a powerful approach to enhance the accuracy and robustness of diagnostic systems. However, effectively integrating multiple sensor data remains a challenge. To address this issue, this paper proposes a novel multi-sensor fusion framework. Firstly, a vibration signal weighted fusion rule based on Kullback-Leibler (K-L) Divergence-Permutation Entropy (PE) is introduced, which adaptively determines the weighting coefficients by considering the positional differences of different sensors. Secondly, a lightweight multi-scale convolutional neural network is designed for feature extraction and fusion of multi-sensor data. An ensemble classifier is employed for fault classification, and an improved hard voting strategy is proposed to achieve more reliable decision fusion. Finally, the superiority of the proposed method is validated using modular state detection data from the Kaggle database.

show abstract

Section: Mbconv Blockmentioning

confidence: 99%

A novel multi-sensor hybrid fusion framework

Du,

Wang,

Zhang

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…last step is a classification layer composed of a 1 � 1 CL, a pooling layer, and a fully connected (FC) layer (for details on the mentioned networks, see refs. [118,119]). Figure 4 shows the structure of MBConv, in which the MBConv network includes two 1 � 1 convolution layers, a depth wise separable CL (kernel size is 3 � 3 or 5 � 5), a squeeze-and-excitation (SE) module, a shortcut connection, and a dropout layer.…”

Section: Deep Learning Model For Feature Extractionmentioning

confidence: 99%

A novel self‐supervised ensemble learning framework for land use and land cover classification of polarimetric synthetic aperture radar images

Darvishnezhad,

Sebt

2023

IET Radar Sonar & Navi

View full text Add to dashboard Cite

Classification with a few samples of training set has been a longstanding issue in the field of polarimetric synthetic aperture radar (PolSAR) image analysis and processing. Aiming at the small number of training samples of the PolSAR image classification task, a novel Self‐Supervised Ensemble Learning Framework (SSELF) is designed. The designed SSELF can automatically extract PolSAR features conducive to PolSAR image classification with a small number of training samples. In addition, it can significantly decrease the dependence of neural network algorithms on large labelled samples of training set. First, utilise the spatial–polarimetric features of PolSAR data perfectly, the EfficientNet‐B0 is presented and utilised as the main section of the Deep Learning (DL) model to extract DL features of PolSAR data. Then, using an optimisation function that constrains the cross‐correlation matrix of various distortions of each sample to the identity matrix, the designed DL model can obtain the effective features of homogeneous samples gathering and heterogeneous samples separating from each other in a self‐supervised manner. Moreover, following the great success of curriculum learning in the area of machine learning, a novel deep curriculum‐learning model is proposed, entitled Deep Curriculum Learning (DCL), to train the DL network in our self‐supervised model. The proposed DCL utilises the entropy‐alpha target decomposition to estimate the degree of complexity of each PolSAR image patch before applying it to the EfficientNet‐B0. Also, an accumulative mini‐batch pacing function is used to introduce more difficult patches to EfficientNet‐B0 in the training process of the designed self‐supervised model. Furthermore, two EL models, feature‐level and view‐level ensemble, are proposed to increase the feature extraction capability and classification result of PolSAR data by jointly using spatial features at different scales and polarimetric information at different bands. In fact, in the proposed feature‐level ensemble strategy, to improve the classification result, the extracted features of the different scales are used as the input of the designed feature fusion algorithm. Therefore, the extracted features are mapped to the new space with lower dimension. In general, in the proposed view‐level ensemble strategy, first, the Mutual Information (MI) between each feature and the other features calculates, and then based‐on the calculated MI, different group of features are selected as various views. Next, the final classification result of the PolSAR image classification is obtained by using the designed majority vote of the result in each view. It should be noted that unlike existing traditional methods, both amplitude and phase information of SAR images to devolve the training process of the proposed model are used. In addition, all of the parameters of the proposed network are developed to a complex domain. In addition, a complex backpropagation algorithm by using the gradient‐based model is used for training the network. Finally, experimental results on three well‐known PolSAR data sets illustrate that the designed SSELF can extract more discriminant features using the designed DL model and can achieve better classification results than the other six novel DL models in terms of small training samples and adequate labelled samples.

show abstract

“…However, some engines will be idle when the computation does not satisfy the order in which the PWC layer is computed before the DWC layer, or when the PWC accelerator is bypassed. Other designs [10][11][12][13][14] that use a partially pipelined architecture have similar engine idle problems, which reduce resource utilization. In contrast to the pipeline architecture is the single-engine architecture.…”

Section: Introductionmentioning

confidence: 99%

A High-Performance FPGA-Based Depthwise Separable Convolution Accelerator

et al. 2023

View full text Add to dashboard Cite

Depthwise separable convolution (DSC) significantly reduces parameter and floating operations with an acceptable loss of accuracy and has been widely used in various lightweight convolutional neural network (CNN) models. In practical applications, however, DSC accelerators based on graphics processing units (GPUs) cannot fully exploit the performance of DSC and are unsuitable for mobile application scenarios. Moreover, low resource utilization due to idle engines is a common problem in DSC accelerator design. In this paper, a high-performance DSC hardware accelerator based on field-programmable gate arrays (FPGAs) is proposed. A highly reusable and scalable multiplication and accumulation engine is proposed to improve the utilization of computational resources. An efficient convolution algorithm is proposed for depthwise convolution (DWC) and pointwise convolution (PWC), respectively, to reduce the on-chip memory occupancy. Meanwhile, the proposed convolution algorithms achieve partial fusion between PWC and DWC, and improve the off-chip memory access efficiency. To maximise bandwidth utilization and reduce latency when reading feature maps, an address mapping method for off-chip accesses is proposed. The performance of the proposed accelerator is demonstrated by implementing MobileNetV2 on an Intel Arria 10 GX660 FPGA by using Verilog HDL. The experimental results show that the proposed DSC accelerator achieves a performance of 205.1 FPS, 128.8 GFLOPS, and 0.24 GOPS/DSP for input images of size 224×224×3.

show abstract

A high-performance convolution block oriented accelerator for MBConv-Based CNNs

Cited by 4 publications

References 8 publications

A novel multi-sensor hybrid fusion framework

A novel multi-sensor hybrid fusion framework

A novel self‐supervised ensemble learning framework for land use and land cover classification of polarimetric synthetic aperture radar images

A High-Performance FPGA-Based Depthwise Separable Convolution Accelerator

Contact Info

Product

Resources

About