arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Author

Sy Lin

2 papers · 1993–1994

1993: 11994: 1
1993peak 1 in 19931994

Frequent co-authors

Papers

showing 1–2 of 2 RSS

Preliminary evaluation of adaptive histogram equalisation for digital mammography

KJ Kearfott, Sy Lin, JS Yakoubian

Computer Vision and Pattern Recognition · · cited 0× · OpenAlex ↗ · more like this ·

The present research is focused on (he application of adaptive histogram equalisation (AHE) to digital mammography. The effects of varying the enhancement factors and the contextual region size on the output histograms for mammographic phantom and breast images are studied. Phantom images are subjected to observer performance tests and the resulting pseudo-receiver operator characteristic (ROC) curves analysed. This analysis shows that the best performance is achieved with AHE, followed by illuminated films, and nonenhanced digital images, respectively. Subjective evaluation of four breast images by two radiology residents resulted in the observation that AHE may increase the difficulty of detection of some tumours but improve visualisation of others. Performance of the method should be unproved with the availability of a higher spatial resolution, lower noise image capture and display…

COMBINING PROJECTED JACOBI METHOD WITH DUAL PROJECTED PSEUDO-QUASI-NEWTON METHOD FOR MULTICOMMODITY NETWORK FLOW PROBLEMS

Sy Lin, Ch Lin

Transportation · Conference on Decision and Control · · cited 0× · OpenAlex ↗ · more like this ·

In this paper, we present a projected Jacobi method combined with a novel dual projected pseudo quasi Newton method for solving general multicommodity network flow problems (MNFPs), which often arise in transportation systems and data communication networks. The quadratic problems induced from the projected Jacobi method in solving the MNFP are difficult due to their large dimensionality; however, they can be solved very efficiently by the proposed dual projected pseudo quasi Newton method, which takes advantage of the network sparsity. We show the convergence of our method and demonstrate its efficiency by several numerical examples. >