• Korean Journal of Ichthyology
• /
• v.2 no.2
• /
• pp.169-181
• /
• 1990

The development of eggs and larvae, and minute scale-like tubercles on the skin surface of larval Acantlaorhodeus gracilis from Korea were observed. They spawned from late March through the middle of June. A fish spawned at least 5 times and the number of eggs averaged 304 per oviposition. Unfertilized eggs are nearly ovoid-shaped(mean$\pm$SD=$2.09{\pm}0.04\;mm$ in length with range of 2.05 to 2.13 mm and mean$\pm$SD=$1.26{\pm}0.02\;mm$ in breadth with range of 1.24 to 1.30 mm measured for the 50 eggs) and opaque yellow color. The tip of egg membrane at the animal pole side swelled and formed a few hilly projections. The shape of the eggs was just like a loquat. Most of embryos began to hatch out in thirty-eight hours after insemination at $22{\pm}1^{\circ}C$ in water temperature. As regards the morphological characters of the eggs and larval development, Acan. gracilis was similar to Acan. asmussi, Acheilognathus rhombeus, A. longipinnis and Pseudoperilampus typus. The larvae of this species is unique particularly in the following two characters, i.e., 1) scale-like tubercles ellipsoided in a diagonal cross section on the whole body and 2) incessant wiggly movement pattern as that of fly maggot, with the larvae of the above mentioned species. These characters seem to reflect the phylogenetic relationships among acheilognathine fishes. On the other hand, this species and Acan. asmussi are spring-summer spawning bitterlings. And also these species never retard the larval growth in such larval stage as the duration from Stage B to Stage D.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.5
• /
• pp.518-528
• /
• 2014

In this paper, we propose a method to recognize the action direction of human by developing 4D space-time (4D-ST, [x,y,z,t]) features. For this, we propose 4D space-time interest points (4D-STIPs, [x,y,z,t]) which are extracted using 3D space (3D-S, [x,y,z]) volumes reconstructed from images of a finite number of different views. Since the proposed features are constructed using volumetric information, the features for arbitrary 2D space (2D-S, [x,y]) viewpoint can be generated by projecting the 3D-S volumes and 4D-STIPs on corresponding image planes in training step. We can recognize the directions of actors in the test video since our training sets, which are projections of 3D-S volumes and 4D-STIPs to various image planes, contain the direction information. The process for recognizing action direction is divided into two steps, firstly we recognize the class of actions and then recognize the action direction using direction information. For the action and direction of action recognition, with the projected 3D-S volumes and 4D-STIPs we construct motion history images (MHIs) and non-motion history images (NMHIs) which encode the moving and non-moving parts of an action respectively. For the action recognition, features are trained by support vector data description (SVDD) according to the action class and recognized by support vector domain density description (SVDDD). For the action direction recognition after recognizing actions, each actions are trained using SVDD according to the direction class and then recognized by SVDDD. In experiments, we train the models using 3D-S volumes from INRIA Xmas Motion Acquisition Sequences (IXMAS) dataset and recognize action direction by constructing a new SNU dataset made for evaluating the action direction recognition.