GIRIST --- Free Iris Recognition Software

GRUS IRIS TOOL

MAIL: girist@grusoft.com
MSN: gsp.cys@gmail.com
QQ: 304718494

GIRIST (GRUS IRIS TOOL) is a free iris recognition software by GRUSOFT. The performance of Girist is comparable to the best commercial systems.It’s based on the novel Giris SDK.

Some Key Features of GIRIST

Average Decidability > 6.0
Average Correct Recognition Rate >95%
When FAR=0.01%, average FRR=5%
Average Extraction time <0.4s
Match rate>50000 irises/second
Handle noises such as eyelash, eyelids and strong reflections
Upper limit of Rotation Angle: ±15 degrees
No limits on size of iris.
Unlimited database size
Three mode: iris authentication, iris identification and iris library verification
Interactive graphical user interface(Figure 1 )

The performance of GIRIST has been verified by the datasets of CASIA, UBIRIS and MMU IRIS. And the testing CPU is Intel Core 2.0 GHz.

Figure 1 graphical user interface of Girist

Download GIRIST 1.0 free.

Download GIRIST tutorial and use guide free.

Experimental Results

The performance of the GIRIST was tested on 6 datasets, 3 of CASIA, 2 of UBIRIS and MMU. Table 1 lists the detail of these datasets. Table 2 lists the experimental results. As you see, GIRIST is fast and stable. GIRIST can handle noises such as eyelash, eyelids and strong reflections(Figure 2).

Table 1 iris datasets

database	No of Images	No of Classes	Image size	Intra-Class Comparisons	Inter-Class Comparisons	characteristic
CASIA-IrisV3-Interval	2655	396	320*280	18042	7028328	Very good image quality with extremely clear iris texture details
CASIA-IrisV3-Lamp	16213	819	640*480	306218	262538938	Nonlinear deformation due to variations of visible illumination
CASIA-IrisV3-Twins	3183	400	640*480	24756	10103550	The first publicly available twins’ iris image dataset
UBIRIS_1	1214	241	200*150	4634	1377166	to minimize noise factors, specially those relative to reflections, luminosity and contrast,
UBIRIS_2	662	241	200*150	1990	301060	to introduce natural luminosity factor. This propitiates the appearance of heterogeneous images with respect to reflections, contrast, luminosity and focus problems.
MMU1 iris	450	90	320*240	1800	200250

Table 2 Experimental Results

database	CRR	FAR/FRR			decidability	Extraction time(s)	Matching Rate	ERR(SEP)
database	CRR	0.01%	0.1%	1%	decidability	Extraction time(s)	Matching Rate	ERR(SEP)
CASIA-IrisV3-Interval	95.2%	4.8%	2.8%	1.6%	6.26	0.13	114339	0.02(0.417)
CASIA-IrisV3-Lamp	88%	9.2%	5.98%	3.18%	5.3	0.5	111942	0.024(0.42)
CASIA-IrisV3-Twins	90.2%	13.4%	9.09%	5.1%	4.92	0.43	112356	0.03(0.418)
UBIRIS_1	96.3%	2.3%	1.25%	0.65%	8.11	0.07	73551	0.006(0.40)
UBIRIS_2	75%	8.5%	5.04%	2.52%	6.59	0.08	41181	0.02(0.40)
MMU iris	98.9%	9.22%	6.11%	3.67%	5.77	0.1	41625	0.03(0.408)

Fig2 Three eyes with strong noises which recognized by GIRIST.