• Journal of Internet Computing and Services
• /
• v.19 no.6
• /
• pp.41-51
• /
• 2018

Recently, in the field of video surveillance, deep learning based learning method is applied to intelligent video surveillance system, and various events such as crime, fire, and abnormal phenomenon can be robustly detected. However, since occlusion occurs due to the loss of 3d information generated by projecting the 3d real-world in 2d image, it is need to consider the occlusion problem in order to accurately detect the object and to estimate the pose. Therefore, in this paper, we detect moving objects by solving the occlusion problem of object detection process by adding depth information to existing RGB information. Then, using the convolution neural network in the detected region, the positions of the 14 keypoints of the human joint region can be predicted. Finally, in order to solve the self-occlusion problem occurring in the pose estimation process, the method for 3d human pose estimation is described by extending the range of estimation to the 3d space using the predicted result of 2d keypoint and the deep neural network. In the future, the result of 2d and 3d pose estimation of this research can be used as easy data for future human behavior recognition and contribute to the development of industrial technology.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.4
• /
• pp.73-82
• /
• 2024

In 2017, the United Nations reported that the population aged 60 and above was increasing more rapidly than all younger age groups worldwide, projecting that by 2050, the population aged 60 and above would constitute at least 25% of the global population, excluding Africa. The world is experiencing a decline in the rate of increase in the working-age population due to global aging, and the younger generation tends to avoid difficult and challenging occupations. Although theoretically, AI equipped with artificial intelligence can replace humans in all fields, in the realm of practical information security, human judgment and expertise are absolutely essential, especially in ethical considerations. Therefore, this paper proposes a method to retrain and reintegrate IT professionals aged 50 and above who are retiring or seeking career transitions, aiming to bring them back into the industry. For this research, surveys were conducted with 21 government/public agencies representing demand and 9 security monitoring companies representing supply. Survey results indicated that both demand (90%) and supply (78%) unanimously agreed on the absolute necessity of such measures. If the results of this research are applied in the field, it could lead to the strategic development of senior information security professionals, laying the foundation for a new market in the Korean information security industry amid the era of low birth rates and longevity.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.41 no.3
• /
• pp.27-36
• /
• 2004

Detection and recognition of human faces in images can be considered as an important aspect for applications that involve interaction between human and computer. In this paper, we propose a face recognition method using eigenfaces and fuzzy neural networks. The Principal Components Analysis (PCA) is one of the most successful technique that have been used to recognize faces in images. In this technique the eigenvectors (eigenfaces) and eigenvalues of an image is extracted from a covariance matrix which is constructed form image database. Face recognition is Performed by projecting an unknown image into the subspace spanned by the eigenfaces and by comparing its position in the face space with the positions of known indivisuals. Based on this technique, we propose a new algorithm for face recognition consisting of 5 steps including preprocessing, eigenfaces generation, design of fuzzy membership function, training of neural network, and recognition. First, each face image in the face database is preprocessed and eigenfaces are created. Fuzzy membership degrees are assigned to 135 eigenface weights, and these membership degrees are then inputted to a neural network to be trained. After training, the output value of the neural network is intupreted as the degree of face closeness to each face in the training database.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.7 no.6
• /
• pp.1385-1391
• /
• 2012

The authenticity of a document is recognized only when sealing on the document can be recognized to be the same as the original seal of a person who drew up the document. But forge in various ways is prevalent; forge through stamp-copy, forge through photocopying, forge through computer seal sculpture machine, forge through photosensitive resin, etc. Among the methods above, a method of making a forged seal with photosensitive resin panels, which is one of the easiest forging methods, enables making a forged seal the same as the original seal through the work of projecting ultraviolet rays on a resin panel on which negative film is stuck after making it using the original seal. But as research into this reality has not been sufficiently carried on, it is difficult to identify forged sealing through photosensitive resin seals, indeed. For this reason, unless regulations about making forged seals using resin machine for print are enforced, crimes related to forged seals using photosensitive resin are expected to increase in the future. Thus, after stamping 10 resin seals made based on an original seal carved of thick and thin strokes of characters, this paper intends to clarify that there are differences between original sealing and forged sealing through sealing identification method. Consequently, we verified the significant difference in the statistics between original sealing and forged sealing.

• 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.