• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.46 no.3
• /
• pp.86-95
• /
• 2009

SIFT is an algorithm to extract vectors at pixels around keypoints, in which the pixel colors are very different from neighbors, such as vertices and edges of an object. The SIFT algorithm is being actively researched for various image processing applications including 3D image reconstructions and intelligent vision system for robots. In this paper, we implement a hardware to sift feature detection algorithm for real time processing in embedded systems. We estimate that the hardware implementation give a performance 25ms of $1,280{\times}960$ image and 5ms of $640{\times}480$ image at 100MHz. And the implemented hardware consumes 45,792 LUTs(85%) with Synplify 8.li synthesis tool.

• Journal of Advanced Navigation Technology
• /
• v.16 no.2
• /
• pp.394-400
• /
• 2012

SURF is a robust local invariant feature descriptor and has been used in many applications such as object recognition. Even though this algorithm has similar matching accuracy compared to the SIFT, which is another popular feature extraction algorithm, it has advantage in matching time. However, these descriptors do not consider relative location information of extracted interesting points to guarantee rotation invariance. Also, since they use gray image of original color image, they do not use the color information of images, either. In this paper, we propose a method for improving matching performance of SURF descriptor using the color and relative location information of interest points. The location information is built from the angles between the line connecting the centers of interest points and the orientation line constructed for the center of each interest points. For the color information, color histogram is constructed for the region of each interest point. We show the performance of our scheme through experiments.

• Journal of the Institute of Convergence Signal Processing
• /
• v.5 no.3
• /
• pp.181-189
• /
• 2004

This paper describes the method for detecting vehicles in the rear and rear-side at night by using headlight features. A headlight is the outstanding feature that can be used to discriminate a vehicle from a dark background. In the segmentation process, a night image is transformed to a binary image that consists of black background and white regions by gray-level thresholding, and noise in the binary image is eliminated by a morphological operation. In the feature extraction process, the geometric features and moment invariant features of a headlight are defined, and they are measured in each segmented region. Regions that are not appropriate to a headlight are filtered by using geometric feature measurement. In region classification, a pair of headlights is detected by using relational features based on the symmetry of a pair of headlights. Experimental results show that this method is very applicable to an approaching vehicle detection system at nighttime.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.1055-1056
• /
• 2008

In this paper, we survey various Robust Object Recognition Algorithms. One of the core technologies for local feature detector is Scale Invariant Feature Transform. And we compared several algorithms with SIFT based on IPP technology. As a result, the conversion of source codes using IPP is sped up. And this will be more improved recognition speed using SIMD Instructions.

• Journal of Institute of Control, Robotics and Systems
• /
• v.2 no.2
• /
• pp.108-114
• /
• 1996

We present proofs of the stability and convergence of Self-organizing feature map (SOFM) neural network with time-invarient learning rate and binary reinforcement function. One of the major problems in Self-organizing feature map neural network concerns with learning rate-"Kalman Filter" gain in stochsatic control field which is monotone decreasing function and converges to 0 for satisfying minimum variance property. In this paper, we show that the stability and convergence of Self-organizing feature map neural network with time-invariant learning rate. The analysis of the proposed algorithm shows that the stability and convergence is guranteed with exponentially stable and weak convergence properties as well.s as well.

• The Journal of Information Technology
• /
• v.12 no.4
• /
• pp.25-32
• /
• 2009

Iris recognition is a biometric technology which can identify a person using the iris pattern. It is important for the iris recognition system to extract the feature which is invariant to changes in iris patterns. Those changes can be occurred by the influence of lights, changes in the size of the pupil, and head tilting. This paper is appropriate for the embedded environment using local gradient histogram embedded system using iris feature extraction methods have implement. The proposed method enables high-speed feature extraction and feature comparison because it requires no additional processing to obtain the rotation invariance, and shows comparable performance to the well-known previous methods.

• Journal of Information Processing Systems
• /
• v.18 no.6
• /
• pp.794-802
• /
• 2022

The recovery of reasonable depth information from different scenes is a popular topic in the field of computer vision. For generating depth maps with better details, we present an efficacious monocular depth prediction framework with coordinate attention and feature fusion. Specifically, the proposed framework contains attention, multi-scale and feature fusion modules. The attention module improves features based on coordinate attention to enhance the predicted effect, whereas the multi-scale module integrates useful low- and high-level contextual features with higher resolution. Moreover, we developed a feature fusion module to combine the heterogeneous features to generate high-quality depth outputs. We also designed a hybrid loss function that measures prediction errors from the perspective of depth and scale-invariant gradients, which contribute to preserving rich details. We conducted the experiments on public RGBD datasets, and the evaluation results show that the proposed scheme can considerably enhance the accuracy of depth prediction, achieving 0.051 for log10 and 0.992 for δ<1.25³ on the NYUv2 dataset.

• Journal of Korea Multimedia Society
• /
• v.3 no.5
• /
• pp.454-460
• /
• 2000

To retrieve complex data such as images, the content-based retrieval method rather than keyword based method is required. In this paper, we implemented a content-based image retrieval system which retrieves object of user query effectively using invariant moments which have invariant properties about linear transformation like position transition, rotation and scaling. To extract the shape feature of objects in an image, we propose a labeling algorithm that extracts objects from an image and apply invariant moments to each object. Hashing method is also applied to reduce a retrieval time and index images effectively. The experimental results demonstrate the high retrieval efficiency i.e precision 85%, recall 23%. Consequently, our retrieval system shows better performance than the conventional system that cannot express the shale of objects exactly.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.12 no.4
• /
• pp.231-238
• /
• 2017

Detecting small infrared targets from the low-SCR images at a long distance is very hard. The previous Local Contrast Method (LCM) algorithm based on the human visual system shows a superior performance of detecting small targets by a background suppression technique through local contrast measure. However, its slow processing speed due to the heavy multi-scale processing overhead is not suitable to a variety of real-time applications. This paper presents a lightweight real-time small target detection algorithm, called by the Improved Selective Local Contrast Method (ISLCM), to reduce the scale-invariant computational overhead. The proposed ISLCM applies the improved local contrast measure to the predicted selective region so that it may have a comparable detection performance as the previous LCM while guaranteeing low scale-invariant computational load by exploiting both adaptive scale estimation and small target feature feasibility. Experimental results show that the proposed algorithm can reduce its computational overhead considerably while maintaining its detection performance compared with the previous LCM.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.13 no.6
• /
• pp.63-69
• /
• 2013

A feature-based image recognition method, using features of an object, can be performed faster than a template matching technique. Invariant feature-based panoramic image generation, an application of image recognition, requires large amount of time to match features between two images. This paper proposes a speed-up method of feature matching using feature strength information. Our algorithm extracts features in images, computes their feature strength information, and selects strong features points which are used to match the selected features. The strong features can be referred to as meaningful ones than the weak features. In the experiments, it was shown that our method speeded up over 40% of processing time than the technique without using feature strength information.