Image Retrieval Based on the Weighted and Regional Integration of CNN Features

• Kaiyang Liao,
• Bing Fan,
• Yuanlin Zheng,
• Guangfeng Lin,
• Congjun Cao,

Vol. 16, No. 3, March 31, 2022
10.3837/tiis.2022.03.008, Download Paper (Free):

• Image retrieval
• Weighting feature
• Global Feature
• Convolutional Neural Network
• Regional integration

Abstract

The features extracted by convolutional neural networks are more descriptive of images than traditional features, and their convolutional layers are more suitable for retrieving images than are fully connected layers. The convolutional layer features will consume considerable time and memory if used directly to match an image. Therefore, this paper proposes a feature weighting and region integration method for convolutional layer features to form global feature vectors and subsequently use them for image matching. First, the 3D feature of the last convolutional layer is extracted, and the convolutional feature is subsequently weighted again to highlight the edge information and position information of the image. Next, we integrate several regional eigenvectors that are processed by sliding windows into a global eigenvector. Finally, the initial ranking of the retrieval is obtained by measuring the similarity of the query image and the test image using the cosine distance, and the final mean Average Precision (mAP) is obtained by using the extended query method for rearrangement. We conduct experiments using the Oxford5k and Paris6k datasets and their extended datasets, Paris106k and Oxford105k. These experimental results indicate that the global feature extracted by the new method can better describe an image.

Statistics

Cite this article