Segmentation and Enhancement of Fingerprint Images Based on Automatic Threshold Calculations

An Increasing Performance of Fingerprint Image Segmentation Based on Clustering and Algorithms

“…From Table 2, in The SVM [11] and K-Means with a 3-dimensional feature [13], the respective value for the misclassification rate in DB1 is 18.75 percent, 20.28 percent respectively. In addition, the segmentation error rate in MP [17], ATC [19] and our proposed method is 0.28%, 13.31% and 0.30% respectively.…”

“…This study describes the experimental operating platform as follows: host configuration: CPU Intel Core2 Duo at 2.00 GHz, RAM 3.00 GB, runtime environment: Microsoft Visual Studio C++ 2013 with OpenCV library.To better verify our algorithm, the following segmentation methods are used in the experiment: SVM [9], 3-dimensional Kmeans [13], MP [17], ACT [19].These algorithms of segmentation were compared with each other.The results have been tested on the public Fingerprint Verification Competition 2004 dataset [25] which contains 4 databases, namely DBl, DB2, DB3, and DB4, to validate the proposed algorithm. The performance measure uses the number of misclassification as defined by (25).…”

“…For comparison of segmentation performance, existing fingerprint image segmentation algorithms, SVM [11], K-Means with a 3-dimensional feature [13], MP [17], and ATC [19] are implemented.The comparison between the suggested technique and others is shown in Table 2 to evaluate the average segmentation mistake for fingerprint images in separate databases. From Table 2, in The SVM [11] and K-Means with a 3-dimensional feature [13], the respective value for the misclassification rate in DB1 is 18.75 percent, 20.28 percent respectively.…”

“…In the second database, if SVM [11 ], K-Means with a 3dimensional feature [13], MP [17] and ATC [19] offer 34.56 percent, 22.30 percent, 292 percent, 29.79 percent respectively in the segmentation rate, but our suggested technique requires a minimum error rate of 1.66 percent.Likewise, the misclassification rate (1.09%) in the last database is less for the proposed algorithm compared to other techniques in the third database. In the fourth database, SVM [11 ] failed to better classify them forwound and background regions (an error rate of 28.89%).…”

“…Subsequently, the threshold level was applied for segmentation. Das, et al [18] achieved segmentation of fingerprints through computer-based statistics and morphological operations.Abboud, et al [19] provided a fresh statistical computing segmentation technique: mean, variance and consistency characteristics of each block in the fingerprint image based on automatic threshold values and Otsu's method.Finally, by using specific sets of rules based on neighboring regions, the filing of the gaps is applied to remove the noise in certain regions in the foreground or background.…”