Comparison Between Convolutional Neural Network CNN and SVM in Skin Cancer Images Recognition
As compared with benign, the common human malignancy of skin cancer can be diagnosed visually starting from clinical screening and ending with histopathological examination. Accurate automatic classification of skin lesion images is a great challenge as the image features are very close in these im...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
middle technical university
2021-12-01
|
| Series: | Journal of Techniques |
| Subjects: | |
| Online Access: | https://journal.mtu.edu.iq/index.php/MTU/article/view/390 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832595134597824512 |
|---|---|
| author | Zaid Ghazi Hadi Ahmed R. Ajel Ayad Q. Al-Dujaili |
| author_facet | Zaid Ghazi Hadi Ahmed R. Ajel Ayad Q. Al-Dujaili |
| author_sort | Zaid Ghazi Hadi |
| collection | DOAJ |
| description |
As compared with benign, the common human malignancy of skin cancer can be diagnosed visually starting from clinical screening and ending with histopathological examination. Accurate automatic classification of skin lesion images is a great challenge as the image features are very close in these images. In this paper we used two methods for image recognition. The first method was carried out with Convolution neural networks (CNN) that promise to provide a potential classifier for skin lesions. This work presents a dermatologist-level classification of skin cancer by using residual network (ResNet-50) as a deep learning convolutional neural network (DLCNN) that maps images to class labels. It presents a classifier with a single CNN to automatically recognize benign and malignant skin images. As for the second method, we used the Support Vector Machine. Which is a supervised learning algorithm and it is used for classification of data for the different classes based on a separating hyperplane. The network inputs are only disease labels and image pixels. About 320 clinical images of the different diseases have been used to train the CNN. The model performance has been tested with untrained images from the two labels. This model identifies the most common skin cancers and can be updated with a new unlimited number of images. The DLCNN was trained by the ResNet-50 model and it showed good classification of the benign and malignant skin categories. The ResNet-50 as a DLCNN has achieved a significant recognition rate of more than 97% on the testing images, which proves that the benign and malignant lesion skin images are properly classified. Support vector machine (SVM) classifier for the classification of skin cancer, for the feature extraction step achieved 86.9% accuracy. This means in CNN; we had more accuracy with 11%. These results were attained using MATLAB.
|
| format | Article |
| id | doaj-art-97f66217228e46c99596f6dfcad1f496 |
| institution | Kabale University |
| issn | 1818-653X 2708-8383 |
| language | English |
| publishDate | 2021-12-01 |
| publisher | middle technical university |
| record_format | Article |
| series | Journal of Techniques |
| spelling | doaj-art-97f66217228e46c99596f6dfcad1f4962025-01-19T11:07:21Zengmiddle technical universityJournal of Techniques1818-653X2708-83832021-12-013410.51173/jt.v3i4.390Comparison Between Convolutional Neural Network CNN and SVM in Skin Cancer Images RecognitionZaid Ghazi Hadi0Ahmed R. Ajel1Ayad Q. Al-Dujaili2Electrical Engineering Technical College, Middle Technical University, Baghdad, IraqElectrical Engineering Technical College, Middle Technical University, Baghdad, IraqElectrical Engineering Technical College, Middle Technical University, Baghdad, Iraq As compared with benign, the common human malignancy of skin cancer can be diagnosed visually starting from clinical screening and ending with histopathological examination. Accurate automatic classification of skin lesion images is a great challenge as the image features are very close in these images. In this paper we used two methods for image recognition. The first method was carried out with Convolution neural networks (CNN) that promise to provide a potential classifier for skin lesions. This work presents a dermatologist-level classification of skin cancer by using residual network (ResNet-50) as a deep learning convolutional neural network (DLCNN) that maps images to class labels. It presents a classifier with a single CNN to automatically recognize benign and malignant skin images. As for the second method, we used the Support Vector Machine. Which is a supervised learning algorithm and it is used for classification of data for the different classes based on a separating hyperplane. The network inputs are only disease labels and image pixels. About 320 clinical images of the different diseases have been used to train the CNN. The model performance has been tested with untrained images from the two labels. This model identifies the most common skin cancers and can be updated with a new unlimited number of images. The DLCNN was trained by the ResNet-50 model and it showed good classification of the benign and malignant skin categories. The ResNet-50 as a DLCNN has achieved a significant recognition rate of more than 97% on the testing images, which proves that the benign and malignant lesion skin images are properly classified. Support vector machine (SVM) classifier for the classification of skin cancer, for the feature extraction step achieved 86.9% accuracy. This means in CNN; we had more accuracy with 11%. These results were attained using MATLAB. https://journal.mtu.edu.iq/index.php/MTU/article/view/390Convolutional Neural Networks (CNN)skin cancersupport vector machine (SVM)lesion classificationmelanoma classificationDLNN |
| spellingShingle | Zaid Ghazi Hadi Ahmed R. Ajel Ayad Q. Al-Dujaili Comparison Between Convolutional Neural Network CNN and SVM in Skin Cancer Images Recognition Journal of Techniques Convolutional Neural Networks (CNN) skin cancer support vector machine (SVM) lesion classification melanoma classification DLNN |
| title | Comparison Between Convolutional Neural Network CNN and SVM in Skin Cancer Images Recognition |
| title_full | Comparison Between Convolutional Neural Network CNN and SVM in Skin Cancer Images Recognition |
| title_fullStr | Comparison Between Convolutional Neural Network CNN and SVM in Skin Cancer Images Recognition |
| title_full_unstemmed | Comparison Between Convolutional Neural Network CNN and SVM in Skin Cancer Images Recognition |
| title_short | Comparison Between Convolutional Neural Network CNN and SVM in Skin Cancer Images Recognition |
| title_sort | comparison between convolutional neural network cnn and svm in skin cancer images recognition |
| topic | Convolutional Neural Networks (CNN) skin cancer support vector machine (SVM) lesion classification melanoma classification DLNN |
| url | https://journal.mtu.edu.iq/index.php/MTU/article/view/390 |
| work_keys_str_mv | AT zaidghazihadi comparisonbetweenconvolutionalneuralnetworkcnnandsvminskincancerimagesrecognition AT ahmedrajel comparisonbetweenconvolutionalneuralnetworkcnnandsvminskincancerimagesrecognition AT ayadqaldujaili comparisonbetweenconvolutionalneuralnetworkcnnandsvminskincancerimagesrecognition |