Evaluation of feature extraction methods for different types of images

Sabry, Eman S.; Elagooz, Salah S.; El‐Samie, Fathi E. Abd; El‐Bahnasawy, Nirmeen A.; El‐Banby, Ghada M.; Ramadan, Rabie A.

doi:10.1007/s12596-022-01024-6

Cited by 5 publications

(3 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This transparency not only improves model interpretability but also helps in identifying potential biases and ensuring fair and ethical deployment of AI algorithms in various applications. As XAI continues to evolve, it is expected to play an increasingly significant role in promoting accountability and reliability in AI systems (Seckin & Seckin, 2022;Sabry et al, 2023;Shuai et al, 2022).…”

Section: Pp 033-050mentioning

confidence: 99%

Metal Surface Texture Classification with Gabor Filter Banks and Xai

Demircioglu,

Seckin,

Torgersen

et al. 2023

DAAAM International Scientific Book

View full text Add to dashboard Cite

This chapter introduces a novel approach for recognizing metal surfaces processed through various operations such as milling, turning, and grinding. The approach involves feature extraction using Gabor filter banks and the utilization of machine learning techniques to facilitate surface recognition. Furthermore, to enhance the interpretability of the model, Explainable Artificial Intelligence (XAI) principles are applied. This is achieved by calculating SHAP (SHapley Additive exPlanations) values for each extracted feature, shedding light on the decision-making process and improving the system's trustworthiness. The Gabor filter with a theta value of 0.79 radians (45°), sigma 5, and lambda value of 0.5 consistently emerged as the most influential feature across all classes. The results of testing on both the train and test data demonstrate a high accuracy and precision for the machine learning model.

show abstract

Section: Pp 033-050mentioning

confidence: 99%

Metal Surface Texture Classification with Gabor Filter Banks and Xai

Demircioglu,

Seckin,

Torgersen

et al. 2023

DAAAM International Scientific Book

View full text Add to dashboard Cite

show abstract

“…43 Secondly, feature engineering techniques can exhibit discriminative power, meaning they possess distinct values for different classes or patterns within the data. 44 For example, in the case of spectra of IR spectroscopy, calculating the differences of statistical values can help highlight the unique characteristics of different classes, making it easier for the classifier to differentiate between them.…”

Section: Feature Engineering -Statistical Featuresmentioning

confidence: 99%

Enhancing glucose classification in continuous flow hydrothermal biomass liquefaction streams through generative AI and IR spectroscopy

et al. 2023

View full text Add to dashboard Cite

show abstract

“…ORB is a fast binary descriptor based on FAST keypoint detector and binary BRIEF descriptor. These extraction methods focus on some specific image features such as points and edges [7], which makes it expensive to detect TM image similarity comprehensively with several manual descriptors. The great improvement in deep learning, that features which might be omitted by human beings can be extracted efficiently by convolutional kernels, makes introducing computer 'opinions' to the procedure of human judgment on TM image similarity a convincing prospect.…”

Section: Introductionmentioning

confidence: 99%

Hybrid-Margin Softmax for the Detection of Trademark Image Similarity

Wang,

Zheng,

Shan

2024

Applied Sciences

View full text Add to dashboard Cite

The detection of image similarity is critical to trademark (TM) legal registration and court judgment on infringement cases. Meanwhile, there are great challenges regarding the annotation of similar pairs and model generalization on rapidly growing data when deep learning is introduced into the task. The research idea of metric learning is naturally suited for the task where similarity of input is given instead of classification, but current methods are not targeted at the task and should be upgraded. To address these issues, loss-driven model training is introduced, and a hybrid-margin softmax (HMS) is proposed exactly based on the peculiarity of TM images. Two additive penalty margins are attached to the softmax to expand the decision boundary and develop greater tolerance for slight differences between similar TM images. With the HMS, a Siamese neural network (SNN) as the feature extractor is further penalized and the discrimination ability is improved. Experiments demonstrate that the detection model trained on HMS can make full use of small numbers of training data and has great discrimination ability on bigger quantities of test data. Meanwhile, the model can reach high performance with less depth of SNN. Extensive experiments indicate that the HMS-driven model trained completely on TM data generalized well on the face recognition (FR) task, which involves another type of image data.

show abstract

Evaluation of feature extraction methods for different types of images

Cited by 5 publications

References 34 publications

Metal Surface Texture Classification with Gabor Filter Banks and Xai

Metal Surface Texture Classification with Gabor Filter Banks and Xai

Enhancing glucose classification in continuous flow hydrothermal biomass liquefaction streams through generative AI and IR spectroscopy

Hybrid-Margin Softmax for the Detection of Trademark Image Similarity

Contact Info

Product

Resources

About