A Comprehensive Survey on Content-Based Image Retrieval Using Machine Learning

Lande, Milind Vijayrao; Ridhorkar, Sonali

doi:10.1007/978-981-16-6285-0_14

Cited by 4 publications

(1 citation statement)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A growing number of individuals are using various electronic devices and platforms, such as mobile phones, digital cameras, and the internet, as a direct result of technological advancements. As the amount of shared and continuously growing multimedia data increases, researchers will have a tougher time finding or retrieving a relevant image from an archive [1][2][3][4][5]. It is advised to apply Modified color motif co-occurrence matrix for image indexing and retrieval, Image indexing and retrieval in JPEG compressed domain based on vector quantization.…”

Section: Introductionmentioning

confidence: 99%

IRFSC: A high-efficiency hybrid model Image Retrieval Based on Feature Selection Using Continuous Learning Process

Lande

Ridhorkar

2023

Preprint

Self Cite

View full text Add to dashboard Cite

Design of high-efficiency feature representation and ranking models is required for retrieval of images based on colour, texture, shape, and other visual aspects. These models must be able to increase retrieval precision while reducing the amount of error and delay required for ranking procedures. Low complexity models can run more quickly, but they are limited in their retrieval performance because they do not exhibit higher retrieval rates. This essay suggests designing a novel hybrid model for high-efficiency feature selection-based picture retrieval using a continuous learning approach to address these problems. A hybrid Elephant Herding Optimization (EHO) & Particle Swarm Optimization (PSO) layer is used in the model's initial extraction of large feature sets from multimodal images in order to continually maximize inter-class feature variance levels. These ranks are post-processed using an incremental optimization method based on Q-Learning, which supports in the continuous optimization of image data sets. As compared to recently proposed state-of-the-art models, the suggested model is able to preserve reduced delay while improving retrieval accuracy by 0.07%, precision by 10.5%, and recall by 3.60%. As a result, the proposed model can be used for a wide range of real-time use cases.

show abstract

Section: Introductionmentioning

confidence: 99%