Multiple classifiers fusion and CNN feature extraction for handwritten digits recognition

Zhao, Haitao; Liu, Ran

doi:10.1007/s41066-019-00158-6

Cited by 91 publications

(47 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Moreover, we will look to apply the proposed rule learning approach in the context of multi-criteria decision making [46,49,50]. In addition, it is worth to conduct in-depth investigation of granular computing techniques [1,3,27,30,48] to achieve deep learning of fuzzy rules in a multi-granularity manner [24], according to the inspiration of deep neural networks [47].…”

Section: Discussionmentioning

confidence: 99%

Multi-stage mixed rule learning approach for advancing performance of rule-based classification

Liu

Chen

2019

Information Sciences

Self Cite

View full text Add to dashboard Cite

Rule learning is a special type of machine learning approaches, and its key advantage is the generation of interpretable models, which provides a transparent process of showing how an input is mapped to an output. Traditional rule learning algorithms are typically based on Boolean logic for inducing rule antecedents, which are very effective for training models on data sets that involve discrete attributes only. When continuous attributes are present in a data set, traditional rule learning approaches need to employ crisp intervals. However, in reality, problems usually show shades of grey, which motivated the development of fuzzy rule learning approaches by employing fuzzy intervals for handling continuous attributes. While a data set contains a large portion of discrete attributes or even no continuous attributes, fuzzy approaches cannot be used to learn rules effectively, leading to a drop in the performance. In this paper, a multi-stage approach of mixed rule learning is proposed, which involves strategic combination of both traditional and fuzzy approaches to handle effectively various types of attributes. We compare our proposed approach with existing algorithms of rule learning. Our experimental results show that our proposed approach leads to significant advances in the performance compared with the existing algorithms.

show abstract

Section: Discussionmentioning

confidence: 99%

Multi-stage mixed rule learning approach for advancing performance of rule-based classification

Liu

Chen

2019

Information Sciences

Self Cite

View full text Add to dashboard Cite

show abstract

“…In paper [5], the authors, Hui-huang Zhao & Han Liu proposed a framework, involving feature extraction using CNNs and a multi-level fusion of diverse classifiers. By preparing different feature sets from Handwritten Digits MNIST Dataset and using different learning algorithms for classifiers' training, they have designed to increase the diversity among classifiers.…”

Section: Related Workmentioning

confidence: 99%

“…By preparing different feature sets from Handwritten Digits MNIST Dataset and using different learning algorithms for classifiers' training, they have designed to increase the diversity among classifiers. In paper [6], the authors compare the accuracies obtained on the CIFAR-10 dataset, using CNN (for feature extraction) and various machine learning algorithms (such as K-Nearest Neighbor, Support Vector Machine Classifier and Fully-connected neural network classifier). In paper [7], the authors used computer vision to detect pneumonia from the frontal chest X-rays.…”

Section: Related Workmentioning

confidence: 99%

“…In paper [6], the authors compare the accuracies obtained on the CIFAR-10 dataset, using CNN (for feature extraction) and various machine learning algorithms (such as K-Nearest Neighbor, Support Vector Machine Classifier and Fully-connected neural network classifier). In paper [7], the authors used computer vision to detect pneumonia from the frontal chest X-rays. They used CNN's feature extraction to extract the features from theX-rays, followed by classifying them as normal and abnormal chest X-rays with the help of support vector machines.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Comparative Analysis of Machine Learning Algorithms with and without Feature Extraction

Gautam¹

2020

IJMTST

View full text Add to dashboard Cite

Image recognition is one of the core disciplines in Computer Vision. It is one of the most widely researched topics of the last few decades. Many advances in image recognition in the past decade, has made it one of the most efficient and powerful disciplines of all, having its applications in every sector including Finance, Healthcare, Security services, Agriculture and many more. Feature extraction is an integral part of image recognition. It helps in training the model more efficiently and with a higher accuracy, by getting rid of any unwanted or unnecessary features, thus reducing the dimensionality of the input image. This also helps in reducing the computational resources required by the algorithm to train, thus making it affordable for people with low end setups. Here we compare the accuracies of different machine learning classification algorithms, and their training times, with and without using feature Extraction. For the purpose of extracting features, a convolutional neural network was used. The model was trained and tested on the data of 12 classes containing a total of 2,175 images. For comparisons, we chose the Logistic regression, K-Nearest Neighbors Classifier, Random forest Classifier, and Support Vector Machine Classifier.

show abstract

“…CNN has been widely used in many fields. Particularly in the field of recognition, CNN is used for handwritten digit recognition [7,8], speech recognition [9,10], facial expression recognition [11,12], human face recognition [13,14], refrigerator fruit and vegetable recognition [15], verification code recognition [16], traffic sign classification [17] and recognition [18], and so on. In the field of image recognition, the images can be directly made the input of CNN, which reduces the complexity of the experiment.…”

Section: Introductionmentioning

confidence: 99%

A Method Combining CNN and ELM for Feature Extraction and Classification of SAR Image

Wang

Zhang

2019

Journal of Sensors

View full text Add to dashboard Cite

Due to the large number of Sigmoid activation function derivation in the traditional convolution neural network (CNN), it is difficult to solve the question of the low efficiency of extracting the feature of Synthetic Aperture Radar (SAR) images. The Sigmoid activation function in the CNN is improved to be a rectified linear unit (ReLU) activation function, and the classifier is modified by the Extreme Learning Machine (ELM). Finally, in this CNN model, the improved CNN works as the feature extractor and ELM performs as a recognizer. A SAR image recognition algorithm based on the CNN-ELM algorithm is proposed by combining the CNN and the ELM algorithm. The experiment is conducted on the Moving and Stationary Target Acquisition and Recognition (MSTAR) database which contains 10 kinds of target images. The experiment result shows that the algorithm can realize the sparsity of the network, alleviate the overfitting problem, and speed up the convergence speed of the network. It is worth mentioning that the running time of this experiment is very short. Compared with other experiment on the same database, it indicates that this experiment has generated a higher recognition rate. The accuracy of the SAR image recognition is 100%.

show abstract

Multiple classifiers fusion and CNN feature extraction for handwritten digits recognition

Cited by 91 publications

References 58 publications

Multi-stage mixed rule learning approach for advancing performance of rule-based classification

Multi-stage mixed rule learning approach for advancing performance of rule-based classification

Comparative Analysis of Machine Learning Algorithms with and without Feature Extraction

A Method Combining CNN and ELM for Feature Extraction and Classification of SAR Image

Contact Info

Product

Resources

About