Scholarlink

<p>Content-based Image Retrieval (CBIR) has garnered significant attention due to the proliferation of digital images and the need for efficient image management. This survey presents a comprehensive overview of handcrafted and deep CBIR schemes, addressing their frameworks, query formation, feature extraction, database indexing, and distance measures. It highlights critical bottlenecks and challenges while evaluating recent advancements. The paper delves into global and local features, focusing on color, shape, texture, and spatial information. The effectiveness of various indexing techniques, including inverted file, hashing, and latent semantic indexing, deep indexing is discussed. Furthermore, we illustrated recent trends in deep learning with special emphasis on supervision categorization in CBIR domain. Also, analysis of recent datasets and performance evaluation metrics, emphasizing accuracy, computational cost, and memory requirements is included. Finally, we propose potential future research directions to broaden CBIR applicability and impact.</p>

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

A comprehensive survey of content based image retrieval schemes: advancements, challenges, and future directions

  • 期刊论文

  • 发表日期：

    2026年2月20日

查看全文
  • Mukul Majhi,
  • Ajay Kumar Mallick

摘要

Content-based Image Retrieval (CBIR) has garnered significant attention due to the proliferation of digital images and the need for efficient image management. This survey presents a comprehensive overview of handcrafted and deep CBIR schemes, addressing their frameworks, query formation, feature extraction, database indexing, and distance measures. It highlights critical bottlenecks and challenges while evaluating recent advancements. The paper delves into global and local features, focusing on color, shape, texture, and spatial information. The effectiveness of various indexing techniques, including inverted file, hashing, and latent semantic indexing, deep indexing is discussed. Furthermore, we illustrated recent trends in deep learning with special emphasis on supervision categorization in CBIR domain. Also, analysis of recent datasets and performance evaluation metrics, emphasizing accuracy, computational cost, and memory requirements is included. Finally, we propose potential future research directions to broaden CBIR applicability and impact.