Algorithms for Vein Image Enhancement and Matching in the Cloud IoT-Based M-Health Environment

Algorithms for Vein Image Enhancement and Matching in the Cloud IoT-Based M-Health Environment

DOI: 10.4018/978-1-7998-4537-9.ch004
OnDemand PDF Download:
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

In this chapter, the authors have described the methodologies to achieve the objectives of veins image enhancement, feature extractions, and matching with other veins images in the cloud IoT-based m-health environment. The initial steps to propose the algorithms for veins image enhance and feature extractions will have five parts. Once the proposed algorithm is written, the hardware architecture designs of the proposed veins image enhancements and feature extraction algorithm will be described by the authors. The hardware designs are presented in subsequent sections of this chapter. Further, the hardware designs are elaborated in detail for each of the techniques. The presented algorithms are implemented in MATLAB 11.0 software, and these algorithms are simulated and integrated with different veins sample images. The hardware designs of veins image enhancements and feature extractions are implemented using Verilog Hardware Language Description (VHLD), and these implemented results are simulated using MSA (Model-Sim-Altera) for sample images of different types of veins.
Chapter Preview
Top

Overview Of Proposed Veins Image Enhancement, Feature Extraction, And Hardware Design Methodologies

See Figure 1.

Figure 1.

An overview of proposed veins image enhancement, feature extraction, and hardware design methodologies

978-1-7998-4537-9.ch004.f01
Top

The Algorithm

It is a well-known fact that the veins exist under the human skin, and veins images are captured online by using an infrared (IR) camera using the trans-illumination method. Sometimes the quality of captured veins images becomes poor because of unusual light scattering and natural human skin absorption characteristics. Hence, veins images may include noises, poor quality illumination and shading artifacts. Thus, it becomes very difficult to differentiate between veins and non-veins. In these situations, the simple technique cannot differentiate whether it is a vein or a non-vein. Hence, veins image improvement becomes important for veins image feature extraction. Thus, an algorithm is to be developed to improve the veins' images and this algorithm can be used to extract the feature of veins pattern in hand veins, palm veins, finger veins, head veins, palm-dorsa hand veins.

The flowchart of figure 2 represents veins image enhancement and feature extraction algorithm, which can be used for extracting veins images from palm veins, hand veins, finger veins, head veins, and palm dorsa veins. The algorithm of figure 2 can be implemented using MATLAB 11.0 or other versions.

Figure 2.

(a) Computing the total number of pixels in veins’ image and extracting the veins’ images; (b) algorithm for thinning veins’ images; (c) the flowchart of the veins’ image enhancement and feature extraction for palm veins, hand veins, finger veins, head veins, and palm-dorsa veins images.

978-1-7998-4537-9.ch004.f02

The Input Image

The veins’ images used in this book for research are given in (Yusoff et al., 2009). The original veins images were captured by a webcam which produced a 640x480 pixel RGB(Red Green Blue) image in JPEG format. Further, the vein images were cropped to 380x290 pixel size by removing the remaining part of the original image and then GIMP (https://www.gimp.org/) software tool was used to convert to the standard size 8-bits gray scale veins image (Yusoff et al., 2009).

Complete Chapter List

Search this Book:
Reset