Scholarlink

Humans have a long history of being susceptible to skin disorders, and now millions of individuals suffer from a wide range of skin conditions. In addition to causing low self-esteem and mental anguish, several of these illnesses are associated with an increased chance of developing skin cancer. Due to the lack of optical resolution for skin disease photos, a medical specialist and sophisticated equipment are required for a proper diagnosis of these conditions. CNN architecture and three preconfigured models(AlexNet, ResNet, and InceptionV3) are part of the proposed deep learning system. For the purpose of Skin Disease Classification, a Dataset of photos featuring seven disorders has been collected. Melanoma, nevi, seborrhoea keratosis, and other skin cancers and benign growths are among them. Since most pre-existing systems categorised cuts and burns as skin diseases, we expanded the dataset to include such photos. Deep Learning algorithms have reduced the requirement for human labour in areas like extracting the features and data restoration for categorization.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Detection and Classification of Skin Disease Using CNN

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • J. Jeyalakshmi,
  • M. Santhiya,
  • M. Shobana

摘要

Humans have a long history of being susceptible to skin disorders, and now millions of individuals suffer from a wide range of skin conditions. In addition to causing low self-esteem and mental anguish, several of these illnesses are associated with an increased chance of developing skin cancer. Due to the lack of optical resolution for skin disease photos, a medical specialist and sophisticated equipment are required for a proper diagnosis of these conditions. CNN architecture and three preconfigured models(AlexNet, ResNet, and InceptionV3) are part of the proposed deep learning system. For the purpose of Skin Disease Classification, a Dataset of photos featuring seven disorders has been collected. Melanoma, nevi, seborrhoea keratosis, and other skin cancers and benign growths are among them. Since most pre-existing systems categorised cuts and burns as skin diseases, we expanded the dataset to include such photos. Deep Learning algorithms have reduced the requirement for human labour in areas like extracting the features and data restoration for categorization.