Shun-ichi Iisaku

Representation and retrieval of video scene by using object actions and their spatio-temporal relationships

In this paper we present a method for representing and retrieving video sequences based on the domain-specific behavior of objects present in the scenes. The representation includes three parts: (a) Action description, representing the action performed by a single object. (b) Interaction description, which describes interactions between multiple objects and is mapped directly to the event semantics in the content domain. (c) Event Structure, which provides a set of spatial and temporal relationship functions, along with a syntax to define the necessary conditions that a particular interaction should meet. Retrieval is performed by processing Event Structures, interpreting object relationships and selecting the relevant combinations of Action descriptions which match the conditions defined in the Event Structures. We describe an implementation of this system for retrieving scenes of soccer plays from among several soccer video sequences.

A visual tracking system for sports video annotation in unconstrained environments

A visual tracking system is presented in which a combination of techniques is used to obtain motion features of objects from a video sequence. Further processing of the motion features gives the spatio-temporal trajectories of the objects, that can be used as cues for annotation. The system solves problems found in tracking objects in unconstrained environments, such as in sports games, where there are multiple objects in motion, the camera performs pan, tilt and zoom movements, and there are objects other than the players in the background. Coarse segmentation is performed with multi-class statistical color models, constructed from samples of the representative colors of each team. Motion vectors are computed to find region correspondence between consecutive frames. Background elements are eliminated by using camera motion parameters, and other false matches are detected by analyzing motion pattern consistency. Finally, objects are registered by placing windows centered in each tracked region. An experimental realization was used to test the system for tracking players in a soccer game, but it could had also been used for generating annotation cues of videos from other sports as well.

Ghost error elimination and superimposition of moving objects in video mosaicing

The paper presents an approach for region based video mosaicing, treating moving objects separately from the background, and with improved ghost-like noise elimination. The mosaic images show the moving objects superimposed over a stationary background. Conventional technologies can reduce the ghost-like noise that occurs from moving objects by using temporal median filtering, but its efficiency depends on the ratio between the speeds of the camera and the moving object. Our technology eliminates these noises more efficiently by using segmented images of a spatio-temporal video sequence. Segmentation is performed using a novel technique that uses different configurations of quad-trees for the initial separation in the split-and-merge process. The segmented images are also used to display tracked moving objects on the panoramic image.

Detection of spicules in mammograms

The objective of this paper is to propose a method to automatically detect spicule shadows in mammograms. The method is consisted of two steps, enhancement and feature selection. First, spicule shadows are enhanced by using a newly developed operation. An opening operation is applied to remove noises and a direction map is made for feature selection. Second, a concentration expression is given with gray levels and two features are selected for recognition of tumors with spicules. In the method, the direction of spicules is not only considered, but the density is also utilized for classification of tumors with spicules. The method was tested on 24 samples including seven tumors with spicules. The recognition rate for tumors with spicules was 100% without the false positives.

Automatic Recognition of Spicules in Mammograms

This paper presents a method of automatic recognition of spicules in mammograms. The method is consisted of two steps, enhancement and feature selection. First, spicule shadows are enhanced by using a newly developed operation. An opening operation is applied to remove noises and a direction map is made for feature selection. Second, a concentration expression is given with gray levels and two features are selected for recognition of tumors with spicules. In the method, the direction of spicules is not only considered, but the density is also utilized. The method was tested on 24 samples including seven tumors with spicules. The recognition rate for tumors with spicules was 100% without the false positives.

Support system for lung cancer screening by CT

This paper presents a system for CAD of lung cancer screening by CT. The system consists of tow parts, automatic processing part and image based diagnosis part. The automatic processing part is to automatically detect the candidate regions of lung cancer based on the methods we proposed, and the image based diagnosis part is mainly used for doctor to make the mass screening. The result obtained by automatic processing part are provided to image based diagnosis pat as support information for increasing the performance of doctor screening.