MICHE I - Mobile Iris CHallenge Evaluation

MICHE Home

MICHE I

MICHE II

our papers about MICHE

BIPLab

Contact:

BIPLab - Biometric and
Image Processing Lab

University of Salerno,
via Ponte Don Melillo,
84084 Fisciano (SA), Italy

cgaldi@unisa.it

MICHE I Submission
Mobile Iris CHallenge Evaluation - Part I

1. Overview. In order to participate in the MICHE I contest, participants are required to provide:
- a paper describing the proposed algorithm and experimental results obtained;
- the database they used for their experiments.

Registration and paper submission have to be performed online via Pattern Recognition Letters website, according to PRL guidelines. Priority will be given to the papers with high novelty and originality. Database submission has to be performed separately, by providing a web link for download.

2. Registration. Papers must be submitted online via the Pattern Recognition Letters website (http://ees.elsevier.com/patrec/), selecting the choice that indicates this special issue (identifier: MICHE-I). After registration to PRL, each participant will receive an ID code. This code will be used as the name of the submitted database.

2.1. Each participant is allowed to submit one single paper and database.

3. Paper Submission. Proposed works may concern but are not limited to the following topics:

- Iris detection on mobile devices;
- Iris segmentation on mobile devices;
- Iris biometric recognition on mobile devices;
- Applications of iris biometrics on mobile devices.

Prepare your paper following the Journal guidelines for Authors
( http://www.elsevier.com/wps/find/journaldescription.cws_home/505619/authorinstructions), which include specifications for submissions aimed at Special Issues. In particular, a maximum of 7500 words is admitted for special issue papers, without counting the References (plus at most 10 Figures/Tables in total).

4. Database Submission. The database must contain close-up iris images acquired with a mobile device (smartphone or tablet). Each image has to be enrich with metadata describing:

- Image filename;
- Image type;
- Left or right iris, or both irises;
- Distance from device;
- Session number;
- Image number;
- User ID;
- User age.
- User gender;
- User ethnicity;
- Device type;
- Device name;
- Camera type;
- Camera name;
- Camera resolution;
- Camera dpi;
- Indoor or outdoor location;
- Artificial, natural or mixed illumination.

      We strongly encourage to provide metadata using the following xml structure (an xml file for each image):

<img>
   <filename>img.jpg</filename>
   <img_type>[iris, face]</img_type>
   <iris>[left, right, both]</iris>
   <distance_from_the_device>[100cm, 10cm]</distance_from_the_device> (in centimeters)
   <session_number>[01, 02, 03, �]</session_number>
   <image_number>[1, 2, 3, 4, ...]</image_number>
   <user id=�[001, 002, �, 022, ...]�>
        <age>[20, 32, 55, ...]</age>
        <gender>[M, F]</gender>
        <ethnicity>[afro american, asian, caucasian, indians, latinos]</ethnicity>
</user>
<device>
      <type>[Smartphone, Tablet]</type>
      <name>[IPhone5, Galaxy S4, Galaxy Tab 2, ...]</name>
      <camera type=�[front, rear]�>
            <name>[VGA, CMOS, iSight, ...]</name>
            <resolution>[0.3MP, ...]</resolution>
            <dpi>[72, ...]</dpi>
     </camera>
</device>
<condition>
     <location>[indoor, outdoor]</location>
     <illumination>[artificial, natural, both]</illumination>
</condition>
<author>[BIPLab, University of ...]</author>
</img>

See an example of metadata association in the publicly available dataset http://biplab.unisa.it/MICHE/database/

4.1. The archive file containing the database has to be named with the ID code obtained after registration on Pattern Recognition Letters.

4.2. The database web link must be sent to cgaldi@unisa.it