Datatsets

Please email me if you are interested to obtain the following databases

Multi-Angle Sclera Dataset (MASD) version 1 [C12, 14, 16, 22]: The dataset consists of 2624 RGB images taken from 82 identities. The images were collected from both the eyes of each individual so 164 different eyes were available. Here for each individual image, four multi-angles (looking straight, left, right and up) are considered and for each angle 4 images are considered. The individuals are comprised of both male and females with different skin complexions, and a few of them were wearing contact lens and images were taken at different times of the day. The database contains images with blinking eyes, closed eyes and blurred eye images. High-resolution images are provided in the database (300 dpi resolution and 7500 x 5000 dimensions). All the images are in JPEG format. A NIKON D 800 camera and 28300 lenses were used for image capturing. Manually annotated segmentation mask of the sclera, iris and the periocular region is available.

Mobile Sclera Dataset (MSD) [C22]: The database consists of 400 RGB images from both eyes of 25 individuals (in other words 50 different eyes). For each eye, 8 sample images were captured. The database contained blurred images and images with blinking eyes. The individuals were comprised of both males and females (12 males and 13 females) of different ages and different skin colours, 2 of them were wearing contact lenses and the images were taken at different times of the day. Variations in image quality (blur, lighting condition etc.) and different acquisition conditions were included intentionally in the database to investigate the performance of the framework in non-ideal scenarios. High-resolution images (3264 × 2448) of 96 dpi are included in the database. All the images are in JPEG format. The images were captured using a mobile camera with an 8-megapixel rear camera. Manually annotated segmentation mask of the sclera region is available.

Sclera Liveness Dataset [J3]: This database consists of 500 genuine RGB images from both eyes of 25 individuals (in other words 50 different eyes). For each eye, 10 sample images were captured. The database contained blurred images and images with blinking eyes. The individuals were comprised of both males and females (12 males and 13 females), of different ages and different skin colours, 2 of them were wearing contact lenses and the images were taken at different times of the day. Variation in image quality (blur, lighting condition, etc.) and different acquisition conditions were included intentionally in the database to investigate the performance of the framework in non-ideal scenarios high-resolution images (3264 × 2448) of 96 dpi are included in the database. All the images are in JPEG format. We have used images of different quality. The images were captured using a mobile camera with an 8 megapixel rear camera. The dataset also consists of 500 fake images. The fake images were prepared automatically by displaying/printing the acquired images of genuine image enrolment and subsequently, a fake image was generated by capturing the displayed image via a camera positioned in front of the screen/printed image. Varying types of display screens were used to produce a real-life situation of a cross sensor scenario, and the same standard sensor was used to acquire them. Few fake images were also developed from the printed eye images.

Face images captured from the optical phenomenon [C17]: Scene images were captured in the wild: 184 face images captured while individual is behind the glass, 74 face images from the reflection on the wall, and 153 face images from the reflection on the glass. Face images from the reflection from different types of the glass wall, shiny wall and glasses were considered while developing the database

Thai student signatures[C21, J5]: Both genuine and forged Thai student signatures were obtained from 100 volunteers. In total, there are 3,000 (100 signer’s × 30 times) genuine signatures obtained. For each of the genuine signatures, 12 skilfully and 12 simple forged signatures were produced; therefore, there are 24 forged signatures per each genuine signer. In total, there are 1,200 (100 signer’s× 12 times) skilfully forged, and 1,200 (100 signers × 12 times) simple forged signatures. Altogether, there are 5,400 signatures in this dataset. Skilled forged signers were asked to learn to forge genuine signatures of the other genuine signers. By learning, they tried to copy the genuine signatures (practised signing the genuine signatures) until they felt confident in forging the genuine signatures. Once they were confident, they provided skilled forged signatures to the collector. Simple forgeries are a set of 12 signatures per user with similar vocal outcomes as the original. The forger only knew how the name sounded and had to decide how to sign the given name as the signature can be signed in a number of ways e.g. signed in Thai, English, Mixed, or signed by just simply writing down the name (so more like writing rather than signing).It was very less occasion when the signature of a simple forger was very close to the genuine signature. It was found that 31 volunteers signed their signatures in English script, whereas the other 64 signed their signatures in Thaiand 5 signers used both scripts to sign a single signature. All samples were scanned at 300 dpi and are binarised. Student signatures, characteristics and their examples can be seen in Figure 1.

Thai name components[c19, 21], there are 6,000 (100 students × 2 name components × 30 times) genuine name components obtained. For each of the genuine name components, 12 skilfully forged name components were produced. In total, there are 2,400 (100 students × 2 name components × 12 times) skilfully forged. Altogether there are 8,400 name components in this dataset. The Thai name components [9], both genuine and forged, were obtained from 100 students, whose ages were between 12 and 16 years old. Each student was asked to write their name (first and last name) 30 times, using the motion time interval technique (described under Thai student signatures Sub-section), on white paper in the given space. All samples were scanned at 300 dpi and are binarised.