Patent
Image classification method based on convolution neural network
Wang Lei,He Yueshun,Wang Kun,Jiang Niande,Zhong Guoyun,Cai Youlin +5 more
- 10 Nov 2017
54
TL;DR: Zhang et al. as mentioned in this paper revealed an image classification method based on a convolution neural network, which comprises the following steps: constructing a deep CNN, improving the CNN, training and testing, and optimizing the CNN parameters.
read more
Abstract: The invention discloses an image classification method based on a convolution neural network. The method comprises the following steps: constructing a deep convolution neural network; improving the deep convolution neural network; training and testing the deep convolution neural network; and optimizing the network parameter. By using the image classification method disclosed by the invention, the improvement and the optimization are respectively performed on the network structure and multiple parameters of the convolution neural network, the recognition rate of the deep convolution neural network can be effectively improved, and the accuracy of the image classification is improved.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Patent
TensorFlow-based deep learning image classification and application deployment method
Chen Changbao,Li Deren,Hou Changsheng,Guo Zhenqiang,Yun Gang,Lu Jianwei +5 more
- 22 Jun 2018
TL;DR: In this article, a TensorFlow-based deep learning image classification and application deployment method is described, which comprises the following steps: 1) constructing Tensorflow machine learning development environment; 2) data acquisition and conversion: obtaining a lot of image data having tags or having no tags form an internet through a distributed crawler system, and carrying out preprocessing on the image data; 3) model establishing and training: establishing a classification model, carrying out training, test and verification on the classification model according to the obtained image data, and keeping the trained classification model; 4
10
Patent
Optimization method and device of convolutional neural network (CNN) and computer storage medium
Zhou Junyang,Wang Xinbo,Ruan Zhifeng,Chen Shuyi,Yu Dahai +4 more
- 08 May 2018
TL;DR: In this paper, an optimization method and device of a convolutional neural network (CNN) and a computer storage medium is presented, in which the CNN includes at least four network layers of an image input layer, at least one CNN layer, one pooling layer and one fully connected layer.
9
Patent
D-MobileNet-based neural network image classification method
TL;DR: Wang et al. as discussed by the authors proposed a method based on Dilated-Mobilenet (DmobileNet) neural network image classification method by combining hole convolution with MobileNet and improving the convolution kernel receptive field of high-resolution input layer.
8
Patent
Article classification method based on depth recovery information
Hou Chunping,Guan Dai,Yang Yang,Lang Yue,Zhang Hengguang +4 more
- 11 Sep 2018
TL;DR: In this paper, the authors proposed a depth recovery model for article classification and monocular image depth estimation in the field of computer vision, which only needs an RGB image instead of a real depth image acquired by a sensor as input.
7
Patent
Image classification method and device, electronic equipment and storage medium
Su Chi,Li Kai,Liu Hongye +2 more
- 30 Nov 2018
TL;DR: In this article, a pre-trained convolutional neural network (CNN) model was used to classify images into categories according to a size relationship between a preset confidence coefficient threshold value of each category and the confidence coefficient indicating that the to-be-classified image belongs to each category.
7
References
Patent
Large-scale face recognition method based on depth convolution neural network model
Wang Zhanxiong,Shao Weiyuan,Feng Rui +2 more
- 20 Jun 2017
TL;DR: Zhang et al. as mentioned in this paper proposed a residual error learning depth network model for large-scale face recognition using a depth convolution neural network (DCNN) model, which is formed by a convolution layer, residual error layer and a full connection layer.
21
Patent
Hyperspectral haze monitoring method based on depth residual error network
Li Yuanxiang,Lu Yongshuai,Shi Yuzhou,Jun Xu,Peng Xishuai +4 more
- 14 Dec 2016
TL;DR: In this article, a hyperspectral haze monitoring method based on a depth residual error network was proposed, and a shortcut passage method was added in the network, so that the training difficulty is lowered, the training precision is improved, a relatively accurate haze monitoring model was obtained, and the monitoring precision was improved.
10
Fast deep neural network based on intelligent dropout and layer skipping
Asma ElAdel,Ridha Ejbali,Mourad Zaied,Chokri Ben Amar +3 more
- 01 May 2017
TL;DR: A fast DCNN based on smart dropout and layer skipping that is possible to classify the image using efficient units of earlier layer(s) and skipping all the subsequent hidden layers directly to the output layer.
10
Patent
Method for recognizing face emotion by using integrated convolutional neural network
Wen Guihua,Hou Zhi +1 more
- 22 Mar 2017
TL;DR: In this paper, a method for recognizing face emotion by using an integrated convolutional neural network (CNN) was proposed, which mainly comprises the following steps: 1, randomly generating a large amount of CNN models by using a method of progressively decreasing each layer of parameter to limit a parameter range; 2, training each CNN model generated randomly to obtain a classifier; 3, calculating the accuracy of each classifier on a validation set; 4, selecting m classifiers having the highest accuracy; and 5, fusing the m classifier by using Bayesian fusion algorithm.
8