Patent
Image classification method based on convolution neural network
Gong Jianrong,Cao Dongxu,Du Kun +2 more
- 21 Dec 2016
9
TL;DR: In this article, a feature filtering layer is added in a hidden layer of the convolutional neural network to reduce the overfitting possibility of the neural network and the training time is reduced.
read more
Abstract: The invention relates to an image classification method based on a convolution neural network. The convolution neural network structure is improved. A feature filtering layer is added in a hidden layer. A large number of features are filtered by extracting the convolution network. A part of features with more noise are removed. The convolution network training efficiency is improved. The training time is reduced. The memory use requirement is reduced. A variety of training techniques are gathered, so that the convolution neural network training is converged a better solution to prevent training parameters from falling into a local minimum area. By reducing the number of network parameters, the over-fitting possibility of the neural network is reduced. The image classification accuracy and efficiency can be effectively 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
Multipath convolutional neural networks-based image classification method and system
Wang Yu,Zhu Ting,Zhang Na,Xiao Hongbing +3 more
- 03 Nov 2017
TL;DR: In this article, a multipath convolutional neural networks-based image classification method and a system is presented, which comprises the steps of inputting a to-be-classified image, partitioning the to be-classified images, and designing a multi-path CNN model, wherein the multipath CNN model at least comprises local feature extraction paths and global feature extraction path.
8
Patent
Image processing method and device and terminal
Zhang Zhiwei,Yang Fan +1 more
- 08 May 2018
TL;DR: In this article, the authors provided an image processing method and device and a terminal, wherein the method comprises the steps of during a process of using a convolutional neural network to carry out the convolutionAL processing on an image, determining whether a currently pre-called first convolutionallayer is equipped with a first chip selection module; if the first convallayer equipped with the first chip-selection module, separately inputting the output data of the previous convolution-layer in the first-chip selection module and the firstconvolutional layer; and calling the first
6
Patent
Sensitive image recognition method and device
TL;DR: In this article, a sensitive image recognition method and device is presented, which includes a convolution neural network based on the sensitive image to obtain the first detection model, wherein the first module is used for determining whether a checked image is a sensitive and location of a sensitive region in the checked image; a level for determining the location of the sensitive region is removed from the first detector and classification training is carried out to generate a second detector.
5
Patent
Picture processing method and device
Xu Zhenqi,Wang Changhu +1 more
- 16 Nov 2018
TL;DR: In this article, the authors present a picture processing method for pre-trained classification models, in which the inconsistency between an object contained in the target picture and a classification result outputted by the classification model is determined by searching for a similar picture of the target image in a training sample set of the model.
5
Patent
Image processing method, device and mobile terminal based on convolutional neural network
Li Qidong,Li Zhiyang,Zhang Wei,Fu Songlin,Gong Qiutang +4 more
- 01 Dec 2017
TL;DR: In this paper, a device and a mobile terminal based on a convolutional neural network (CNN) are presented, which is suitable for the execution in mobile terminal with a graphic program interface.
5
References
•Proceedings Article
ImageNet Classification with Deep Convolutional Neural Networks
Alex Krizhevsky,Ilya Sutskever,Geoffrey E. Hinton +2 more
- 03 Dec 2012
TL;DR: The state-of-the-art performance of CNNs was achieved by Deep Convolutional Neural Networks (DCNNs) as discussed by the authors, which consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully-connected layers with a final 1000-way softmax.
•Posted Content
Improving neural networks by preventing co-adaptation of feature detectors
TL;DR: The authors randomly omits half of the feature detectors on each training case to prevent complex co-adaptations in which a feature detector is only helpful in the context of several other specific feature detectors.
Patent
Automatic insect image identification method based on depth convolutional neural network
Liu Ziyi,Yong He,Yang Guoguo +2 more
- 19 Aug 2015
TL;DR: In this article, an automatic insect image identification method based on a depth convolutional neural network (DCNN) was proposed. But the method is not suitable for the detection of the specific types of insects.
63
•Posted Content
Deep Temporal Appearance-Geometry Network for Facial Expression Recognition.
TL;DR: A deep learning technique which is regarded as a tool to automatically extract useful features from raw data, is adopted and one of the main contributions of this paper is that the deep network catches the facial action points automatically.
59
Patent
Breast cancer cell characteristic analysis system based on deep learning
Guo Yanyan,Liu Da,Liu Kui,Hu Piao +3 more
- 30 Mar 2016
TL;DR: In this article, a breast cancer cell characteristic analysis system based on deep learning is presented, in which a multilevel convolutional neural network is used to extract multi-level features and an activation function of a model is an unsaturated ReLU function with faster convergence properties.
22