• Journal of the Semiconductor & Display Technology
• /
• v.22 no.2
• /
• pp.128-132
• /
• 2023

Recently, with the development of deep learning technology, research on pedestrian attribute recognition technology using deep neural networks has been actively conducted. Existing pedestrian attribute recognition techniques can be obtained in such a way as global-based, regional-area-based, visual attention-based, sequential prediction-based, and newly designed loss function-based, depending on how pedestrian attributes are detected. It is known that the performance of these pedestrian attribute recognition technologies varies greatly depending on the type of backbone network that constitutes the deep neural networks model. Therefore, in this paper, several backbone networks are applied to the baseline pedestrian attribute recognition model and the performance changes of the model are analyzed. In this paper, the analysis is conducted using Resnet34, Resnet50, Resnet101, Swin-tiny, and Swinv2-tiny, which are representative backbone networks used in the fields of image classification, object detection, etc. Furthermore, this paper analyzes the change in time complexity when inferencing each backbone network using a CPU and a GPU.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.16 no.5
• /
• pp.171-177
• /
• 2021

It is difficult to accurately extract features of pedestrian because the pedestrian is photographed at a crosswalk using a camera positioned higher than the pedestrian. In addition, it is more difficult to extract features when a part of the pedestrian's body is covered by an umbrella or parasol or when the pedestrian is holding an object. Representative methods to solve this problem include Object Detection, Instance Segmentation, and Pose Estimation. Among them, this study intends to use the Pose Estimation method. In particular, we intend to increase the recognition rate of pedestrians in crosswalks by maintaining the image resolution through HigherHRNet and applying the foot estimation technique. Finally, we show the superiority of the proposed method by applying and analyzing several data sets covered by body parts to the existing method and the proposed method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.9
• /
• pp.3762-3781
• /
• 2020

Infrared pedestrian target recognition is a vital problem of significant interest in computer vision. In this work, a novel infrared pedestrian target recognition method that uses heterogeneous features with multi-kernel transfer learning is proposed. Firstly, to exploit the characteristics of infrared pedestrian targets fully, a novel multi-scale monogenic filtering-based completed local binary pattern descriptor, referred to as MSMF-CLBP, is designed to extract the texture information, and then an improved histogram of oriented gradient-fisher vector descriptor, referred to as HOG-FV, is proposed to extract the shape information. Second, to enrich the semantic content of feature expression, these two heterogeneous features are integrated to get more complete representation for infrared pedestrian targets. Third, to overcome the defects, such as poor generalization, scarcity of tagged infrared samples, distributional and semantic deviations between the training and testing samples, of the state-of-the-art classifiers, an effective multi-kernel transfer learning classifier called MK-TrAdaBoost is designed. Experimental results show that the proposed method outperforms many state-of-the-art recognition approaches for infrared pedestrian targets.

• Journal of Biomedical Engineering Research
• /
• v.31 no.6
• /
• pp.481-486
• /
• 2010

In this paper, we suggest an advanced algorithm, to recognize pedestrian/non-pedestrian using differential haar-like feature, which applies Adaboost algorithm to make a strong classification from weak classifications. First, we extract two feature vectors: horizontal haar-like feature and vertical haar-like feature. For the next, we calculate the proposed feature vector using differential haar-like method. And then, a strong classification needs to be obtained from weak classifications for composite recognition method using the differential area of horizontal and vertical haar-like. In the proposed method, we use one feature vector and one strong classification for the first stage of recognition. Based on our experiment, the proposed algorithm shows higher recognition rate compared to the traditional method for the pedestrian and non-pedestrian.

• Journal of Korea Multimedia Society
• /
• v.21 no.7
• /
• pp.755-763
• /
• 2018

In this paper, for recognize a night pedestrian from an infrared video, a new method differentiated from the existing feature vector is proposed and experimented. The new approach focuses on the shape feature vector of the structure and shape of the pedestrian image divided by the human body seven split ratio. The pedestrian images are divided into 7 square blocks from the still image of the preprocessing process. And to reduce the dimension, the square block is converted into a mosaic block. The scalar and direction of the shape feature vector is calculated by the brightness and position of the element in the mosaic. For practicality of infrared video system, the proposed method simplifies the data to be processed by reducing the amount of data in the preprocessing in order to continuously batch process the entire system in real time. Through the experiments, we verified the validity of the proposed shape feature vector. In comparison to the existing method, we propose a new shape feature vector generation method as the feature vector for night pedestrian recognition.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.6
• /
• pp.654-662
• /
• 2010

In this paper, we suggest an advanced algorithm, to recognize pedestrian/non-pedestrian using second-stage cascade method, which applies Adaboost algorithm to make a strong classification from weak classifications. First, we extract two feature vectors: (i) Histogram of Oriented Gradient (HOG) which includes gradient information and differential magnitude; (ii) Curvature-HOG which is based on four different curvature features per pixel. And then, a strong classification needs to be obtained from weak classifications for composite recognition method using both HOG and curvature-HOG. In the proposed method, we use one feature vector and one strong classification for the first stage of recognition. For the recognition-failed image, the other feature and strong classification will be used for the second stage of recognition. Based on our experiment, the proposed algorithm shows higher recognition rate compared to the traditional method.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.8 no.2
• /
• pp.121-128
• /
• 2013

There have been various research efforts for pedestrian recognition in embedded imaging systems. However, many suffer from their heavy computational complexities. SVM classification method has been widely used for pedestrian recognition. The reduction of candidate region is crucial for low-complexity scheme. In this paper, We propose a real time HOG based pedestrian detection system on GPU which images are captured by a pair of cameras. To speed up humans on road detection, the proposed method reduces a number of detection windows with disparity-search and near-search algorithm and uses the GPU and the NVIDIA CUDA framework. This method can be achieved speedups of 20% or more compared to the recent GPU implementations. The effectiveness of our algorithm is demonstrated in terms of the processing time and the detection performance.

• Journal of Information Technology Services
• /
• v.15 no.4
• /
• pp.73-83
• /
• 2016

Pedestrian safety was recently highlighted with a yellow carpet, a yellow-colored pavement material prepared for children waiting for signals for pedestrian crossing, without validation of its efficiency in practice. It was a promising device likely to assist highway safety by stimulating pedestrian to step on the yellow-colored area; it was generally called nudge effects. This paper delivers a study conducted to check the effectiveness of yellow carpet in three different aspects in vehicle driver's perspective by applying the newly introduced information technology (IT) service: Visual Attention Software (VAS). It was assumed that VAS developed by 3M in the United States should be able explain the Korean drivers' visual reaction behaviors since technology embedded in VAS was developed based on and proved by other various international countries and continents in the world. A set of pictures was taken at thirteen different field sites in seven school zone areas in the Seoul metropolitan area before and after the installation of a yellow carpet, respectively. Sets of those pictures were analyzed with VAS, and the results were compared based on the selective safety measures: the likely focusing on standing pedestrians (waiting for a pedestrian's green signal time) affected by its background (yellow-colored pavement) contrasting him or her. The test results from a set of before-and-after comparison analyses showed that the placement of yellow carpet would (1) increase 71% of driver's visual attention on pedestrian crossing areas and (2) change the sequential order of visual attention on that area 2.4 steps ahead. The findings would enhance deployment of such promising efficiency and thus increase children safety in pedestrian crossing. The result was promising to highlight the way to support the changes in conservative traffic safety engineering field by applying the advanced IT services, while much robust research was recommended to overcome the limitation of simplification of this study.

• Journal of Electrical Engineering and Technology
• /
• v.13 no.2
• /
• pp.961-968
• /
• 2018

Recently, researches on automatic recognition of human activities have been actively carried out with the emergence of various intelligent systems. Since a large amount of visual data can be secured through Closed Circuit Television, it is required to recognize human behavior in a dynamic situation rather than a static situation. In this paper, we propose new intelligent human activity recognition model using the trajectory information extracted from the video sequence. The proposed model consists of three steps: segmentation and partitioning of trajectory step, feature extraction step, and behavioral learning step. First, the entire trajectory is fuzzy partitioned according to the motion characteristics, and then temporal features and spatial features are extracted. Using the extracted features, four pedestrian behaviors were modeled by decision tree learning algorithm and performance evaluation was performed. The experiments in this paper were conducted using Caviar data sets. Experimental results show that trajectory provides good activity recognition accuracy by extracting instantaneous property and distinctive regional property.

• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.5
• /
• pp.490-498
• /
• 2011

In this paper, we present a two-stage vision-based approach to detect multi views of pedestrian in road scene images. The first stage is HG (Hypothesis Generation), in which potential pedestrian are hypothesized. During the hypothesis generation step, we use a vertical, horizontal edge map, and different colors between road background and pedestrian's clothes to determine the leg position of pedestrian, then a novel symmetry peaks processing is performed to define how many pedestrians is covered in one potential candidate region. Finally, the real candidate region where pedestrian exists will be constructed. The second stage is HV (Hypothesis Verification). In this stage, all hypotheses are verified by Support Vector Machine for classification, which is robust for multi views of pedestrian detection and recognition problems.