Junta Doi

Personal authentication using feature points on finger and palmar creases

A new and practical method of reliable and real-time authentication is proposed. Finger geometry and feature extraction of the palmar flexion creases are integrated in a few numbers of discrete points for faster and robust processing. A video image of either palm, palm placed freely facing toward a near infrared video camera in front of a low-reflective board, is acquired. Fingers are brought together without any constraints. Discrete feature point involves intersection points of the three digital (finger) flexion creases on the four finger skeletal lines and intersection points of the major palmar flexion creases on the extended finger skeletal lines, and orientations of the creases at the points. These metrics define the feature vectors for matching. Matching results are perfect for 50 subjects so far. This point wise processing, extracting enough feature from non contacting video image, requiring no time-consumptive palm print image analysis, and requiring less than one second processing time, will contribute to a real-time and reliable authentication.

Reuse of a geometric model for shape approximation

A robust procedure of three-dimensional shape modeling for an object approximation is proposed. A solid model is first prepared or reused for the prototype and then transferred to the model that approximates the shape of the object. Vertex geometry of the prototype is modified based on the multidirectional silhouettes and light stripe pattern projection. Topology of the mesh, shape and connectivity of the vertices, of the prototype is conserved throughout the modeling process. Stable meshing and accurate shape approximation are achieved, starting from a simple or a dedicated model with desired meshes. A dedicated male sculpture and a simple cylinder with quadrilateral meshes are demonstrated for the shape approximation. This eliminates the laborious modeling procedures, reusing a proven or simple and stable model, and the complicated camera calibrations.

Noise-resistive 3D shape modeling for virtual heritage applications

This paper proposes a practical, topologically robust and ranging error resistive shape modeling procedure that approximates a real 3D object, such as heritage artifacts, with the matrix-format meshing for the 3D shape processing. The processing is used for the modeled shape modification, especially, for restoration of the broken parts of the artifacts or the virtual manipulation of the 3D shape. A geometric model with desired meshing is directly reconstructed based on a solid modeling approach. The radial distance of each scanning point from the axis of the cylindrical coordinates is measured by laser triangulation. The angular and vertical positions of the laser beam are two other coordinate values of the scanning. A face array listing (topology), which defines the vertex (sampling point) connectivity and the shape of the mesh, is assigned to meet the desired meshing. Stable meshing, and hence, an accurate approximation, free from the shape ambiguity unavoidable in the widely used ICP (iterative closest point) modeling, is then accomplished. This proposal allows a versatile and automatic shape reconstruction for cultural heritage retrieval, restoration and virtual training, coping with the unavoidable error problem inherent in the cultural heritage.

Biometric authentication using finger and palmar creases

A practical method for a non-contacting and real-time authentication is proposed. The finger geometry and feature extraction of the palmar flexion creases are integrated into a small number of discrete points. For a video image of either palm, a palm placed freely facing toward a video camera is acquired. The fingers are brought together and the palm is straightened out without any constraints. The discrete feature points involve intersection points of the three finger (digital) flexion creases on the four finger skeletal lines and the intersection points of the major palmar flexion creases or prominent creases on the extended finger skeletal lines and the orientations of the creases at the points. The matching so far results are perfect for about 500 palm samples from 50 subjects. This discrete point processing, requiring no time-consumptive palm print image analysis and requiring less than one second processing time, will contribute to a non-contacting, real-time and reliable authentication.

Discrete finger and palmar feature extraction for personal authentication

A new method of a reliable and real-time authentication is proposed. Finger geometry and feature extraction of the palmar flexion creases are integrated in discrete points of characteristics. A video image of either palm, palm placed freely facing toward a video camera in front of a low-reflective board, is acquired. Fingers are brought together without any constraints. Discrete feature point extraction for each of the four fingers involves: intersection points of the three digital (finger) flexion creases on the finger skeletal line; skeletal lengths of the finger segments between the three creases; distances between the intersection points and the corresponding points of the adjacent fingers. Discrete feature extraction for the palm involves: intersection points of the major palmar flexion creases on the extended finger skeletal line; orientation of the crease at each point of the intersection. These metrics define the feature vectors for matching. Matching results are perfect for 50 subjects so far. This point wise integration of the finger and palmar feature extraction, extracting enough feature from non contacting video image, requiring no time-consumptive palm print image analysis, and requiring less than one second processing time, will contribute to a real-time and reliable authentication.

Personal authentication by integrating palmar geometry and flexion crease analysis

A personal authentication method is proposed by integrating palmar geometry with the palmar and finger flexion crease analysis. A 900 x 900 image of either palm, placed freely on the flat transparent plate, is captured. Feature extraction involves: area, width and perimeter of the palm; areas, perimeters, skeletal axes and their lengths of the four fingers; shape factors of the palm and the fingers derived from the areas and the perimeters; aspect ratios; lengths of all of the finger flexion creases; intersecting points of the finger axes and the finger flexion creases; intersecting points of the finger axes and the major palmar flexion creases, those are prominent and typically classified into the thenar crease, the proximal transverse crease and the distal transverse crease. Some minor or secondary flexion creases are additionally detected. Orientation of the crease at each point of intersection is also detected. These metrics define the feature vectors for matching. We have tested the method on a limited set of palm images collected in a laboratory environment. Matching results, especially featured the oriented intersecting points of palmar creases, are encouraging. This integration with the palmar feature extraction will contribute to a more robust and reliable authentication system.

Solid model approximation for successive three-dimensional shape processing

This paper proposes a practical, accurate, topologically robust and ranging error resistive shape modeling procedure that approximates a real object, with the matrix-format data structure, for the resulting 3D shape processing. Examples of the shape processing are based on the premise of the virtual manipulation of the 3D shape, such as local shape modification and blending. A geometric model with the desired meshing is directly reconstructed based on a solid modeling approach. The radial distance of each scanning point from the axis of the cylindrical coordinates is measured by laser triangulation. The angular and vertical positions of the laser beam are two other coordinate values of the scanning. A face array listing (topology), which defines the vertex (sampling point) connectivity and the shape of the mesh, is assigned to meet the desired meshing. Topologically stable meshing, and hence, an accurate approximation, free from the shape ambiguity unavoidable in the so-called ICP (iterative closest point) modeling, is then accomplished. This proposal allows not only the versatile and automatic shape reconstruction, but also virtual shape manipulation for various trainings and restorations

Three-dimensional shape modeling for distance education

This paper proposes a topologically robust and accurate shape modeling procedure to prepare materials for distance learning. The procedure approximates a real 3D object based not on the recently used so-called ICP (Iterative Closed Point) algorithm or the Delaunay triangulation for noise resistive reconstruction. An easy-to-manipulate geometric model with desired meshing, not limited to a triangular one, but also organized and matrix-formatted quadrilateral, hexagonal or n-gonal mesh, is directly reconstructed based on a solid modeling approach. The radial distance of each scanning point is measured using a laser triangulation sensor. The mesh connectivity is assigned and conserved throughout the process. Examples of archeological replicas and stone specimens are demonstrated.

Topology conserved 3D reconstruction and shape processing for reuse of the geometric models

This paper proposes a practical, topologically robust and noise resistive shape modeling procedure that approximates a real 3D object with the matrix-form organized meshing for the 3D shape processing, focusing on the reuse of geometric models. A geometric model with desired meshing, not limited to triangular one, but also quadrilateral, hexagonal or n-gonal mesh, is directly reconstructed based on a solid modeling approach. The radial distance of each scanning point from the axis of the cylindrical coordinates is measured using a laser triangulation sensor. The angular and vertical positions of the laser beam are two other coordinate values of the scanning. A face array listing (topology), which defines the vertex (sampling point) connectivity and the shape of the mesh, is assigned and conserved to meet the desired meshing. Stable meshing, and hence, an accurate approximation, free from the misconnection unavoidable in the widely used ICP (iterative closest point)-based modeling, is then accomplished. This proposal, sensory-modeling and information fusion, allows a versatile shape approximation, suitable for reuse of the geometric models.

Three-dimensional reconstruction of solid models from multi-directional images and applications to industrial mensuration

An automated modeling system that represents an object body by a solid model of polyhedral approximation was developed using multi-directional image inputs. The principle of this solid model generation is based on the calculation of viewing pyramids consisting of the image boundary and the corresponding focal point. Applications to the non-contacting three-dimensional measurement of the shape of irregular and complicated object such as a doll, a fish, a spiral shell, and a lemon, to the automated mesh generation for the boundary element analysis of the lemon, to the CAD of a tillage blade, and to the shape, deformation and motion analysis of an aerodynamic tuft, a small tracer particle, and an alcohol wick flame for the flow visualization are presented. It appears to have considerable utility in bio-mechanics and various industrial applications.