Improved bag-of-features using grey relational analysis for classification of histology images

Abstract

An efficient classification method to categorize histopathological images is a challenging research problem. In this paper, an improved bag-of-features approach is presented as an efficient image classification method. In bag-of-features, a large number of keypoints are extracted from histopathological images that increases the computational cost of the codebook construction step. Therefore, to select the a relevant subset of keypoints, a new keypoints selection method is introduced in the bag-of-features method. To validate the performance of the proposed method, an extensive experimental analysis is conducted on two standard histopathological image datasets, namely ADL and Blue histology datasets. The proposed keypoint selection method reduces the extracted high dimensional features by 95% and 68% from the ADL and Blue histology datasets respectively with less computational time. Moreover, the enhanced bag-of-features method increases classification accuracy by from other considered classification methods.

Advanced imaging system for brain tumor automatic classification from MRI images using HOG and BOF feature extraction approaches

“…Finally, without a label, test set images are sent to the trained classifier to predict their labels. 16,17…”

“…The GridStep is [8 × 8] and the Block Width is [32 × 64 × 96 × 128], (ii) for the representation of the dictionary, a visual vocabulary is created by clustering features extracted from a training set using the K‐means clustering algorithm, (iii) each image represents a histogram of visual word occurrences contained, and (iv) these histograms and image labels are used to train the classifier. Finally, without a label, test set images are sent to the trained classifier to predict their labels 16,17 …”

Energy Efficient Multi-criterion Binary Grey Wolf Optimizer based Clustering for Heterogeneous Wireless Sensor Networks

“…A number of heuristics and meta-heuristics-based clustering techniques are available in the literature for minimizing energy consumption in sensor networks [14] [15] [16] [17]. Further, meta-heuristic techniques have been largely applied to find optimal solutions to the clustering problems [18] [19] in WSN such as intelligent hierarchical clustering and routing protocol (IHCR) [20], evolutionary routing protocol (ERP) [11], KGA [21], KBBO [22] [23], and many more. Many real-world problems in WSN require optimization of multiple objectives simultaneously, such as longest network lifetime, minimum latency, maximum energy saving, maximum coverage, and connectivity, etc.…”