scholarly journals SCN: A Novel Shape Classification Algorithm Based on Convolutional Neural Network

Symmetry ◽  
2021 ◽  
Vol 13 (3) ◽  
pp. 499 ◽  
Author(s):  
Chaoyan Zhang ◽  
Yan Zheng ◽  
Baolong Guo ◽  
Cheng Li ◽  
Nannan Liao
Keyword(s):  
Network Structure ◽  
Shape Matching ◽  
Scale Space ◽  
Shape Classification ◽  
Shape Context ◽  
Curvature Scale Space ◽  
Classification Tasks ◽  
Basic Network ◽  
Important Branch ◽  
Better Than

Shape classification and matching is an important branch of computer vision. It is widely used in image retrieval and target tracking. Shape context method, curvature scale space (CSS) operator and its improvement have been the main algorithms of shape matching and classification. The shape classification network (SCN) algorithm is proposed inspired by LeNet5 basic network structure. Then, the network structure of SCN is introduced and analyzed in detail, and the specific parameters of the network structure are explained. In the experimental part, SCN is used to perform classification tasks on three shape datasets, and the advantages and limitations of our algorithm are analyzed in detail according to the experimental results. SCN performs better than many traditional shape classification algorithms. Accordingly, a practical example is given to show that SCN can save computing resources.

Image Fusion for Infrared Thermography and Inspection of Pressure Vessel

2009 ◽  
Vol 131 (2) ◽  
Author(s):  
Yu-Peng Tian ◽  
Ke-Yin Zhou ◽  
Xiaowei Feng ◽  
Sheng-Lin Yu ◽  
Hua Liang ◽  
...  
Keyword(s):  
Image Fusion ◽  
Infrared Thermography ◽  
Pressure Vessel ◽  
Evaluation Method ◽  
Scale Space ◽  
Infrared Light ◽  
Shape Context ◽  
Curvature Scale Space ◽  
Fusion Methods ◽  
Space Method

Infrared thermography is a nondestructive evaluation method with an increasing span of applications. In this paper the pixel-level image fusion of infrared light and visible light image for infrared thermography and its application in nondestructive testing (NDT) of pressure vessel are studied. First in the image fusion, different mode image registration has been done based on feature corners. The corners and curves relative to each corner are extracted using the curvature scale space method. A new shape context descriptor for each corner is computed as the criterion to match the corners. Marks can then be added to pick up the corners, especially if there are no distinct matching feature corners between the different mode images. Two fusion methods, contrast modulation and transparent fusion, are introduced to fuse the visual and infrared light images. The experimental results show that the methods are good for pixel-level image fusion. Finally the method is applied to inspection of pressure vessel. The research indicates that image fusion for infrared thermography can facilitate defect location and interpretation of inspection results.

THE NORMALIZED WAVELET DESCRIPTOR FOR SHAPE RETRIEVAL

2008 ◽  
Vol 06 (01) ◽  
pp. 25-36 ◽  
Author(s):  
KIMCHENG KITH ◽  
BAREND J. VAN WYK ◽  
MICHAËL A. VAN WYK
Keyword(s):  
Image Analysis ◽  
Image Retrieval ◽  
Shape Descriptor ◽  
Scale Space ◽  
Experimental Results ◽  
Shape Retrieval ◽  
Fourier Descriptor ◽  
Curvature Scale Space ◽  
Wavelet Descriptor

In many image analysis applications, such as image retrieval, the shape of an object is of primary importance. In this paper, a new shape descriptor, namely the Normalized Wavelet Descriptor (NWD), which is a generalization and extension of the Wavelet Descriptor (WD), is introduced. The NWD is compared to the Fourier Descriptor (FD), which in image retrieval experiments conducted by Zhang and Lu, outperformed even the Curvature Scale Space Descriptor (CSSD). Image retrieval experiments have been conducted using a dataset containing 2D-contours of 1400 objects extracted from the standard MPEG7 database. For the chosen dataset, our experimental results show that the NWD outperforms the FD.

Multi-scale curvature product for robust image corner detection in curvature scale space

2007 ◽  
Vol 28 (5) ◽  
pp. 545-554 ◽  
Author(s):  
Xiaohong Zhang ◽  
Ming Lei ◽  
Dan Yang ◽  
Yuzhu Wang ◽  
Litao Ma
Keyword(s):  
Scale Space ◽  
Corner Detection ◽  
Multi Scale ◽  
Curvature Scale Space ◽  
Robust Image

Optimizing Convolutional Neural Network Accelerator on Low-Cost FPGA

2021 ◽  
pp. 2150193
Author(s):  
Truong Quang Vinh ◽  
Dinh Viet Hai
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Low Cost ◽  
Optimal Number ◽  
Data Reuse ◽  
Logic Element ◽  
Input Buffer ◽  
Classification Tasks ◽  
Processing Engine ◽  
Better Than

Convolutional neural network (CNN) is one of the most promising algorithms that outweighs other traditional methods in terms of accuracy in classification tasks. However, several CNNs, such as VGG, demand a huge computation in convolutional layers. Many accelerators implemented on powerful FPGAs have been introduced to address the problems. In this paper, we present a VGG-based accelerator which is optimized for a low-cost FPGA. In order to optimize the FPGA resource of logic element and memory, we propose a dedicated input buffer that maximizes the data reuse. In addition, we design a low resource processing engine with the optimal number of Multiply Accumulate (MAC) units. In the experiments, we use VGG16 model for inference to evaluate the performance of our accelerator and achieve a throughput of 38.8[Formula: see text]GOPS at a clock speed of 150[Formula: see text]MHz on Intel Cyclone V SX SoC. The experimental results show that our design is better than previous works in terms of resource efficiency.

Sign in / Sign up

Export Citation Format

Share Document