• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.11
• /
• pp.2055-2061
• /
• 2006

The detection of geometric primitives from a digital image is one of the basic tasks in computer vision area and the Hough transform is a well-known method for detecting analytical shape represented by a number of free parameters. However the basic property of the Hough transform, the one-to-many mapping from an image space to a Hough space, causes the innate problem, the sensitivity to noise. In this paper, we proposed Edge Strength Hough Transform which uses edge strength to reduce the sensitivity to noise and proved the insensitivity using the ratio of peaks in a Mough space. We also experimented the proposed method on lines and got small number of peaks in a Hough space compared to traditional Hough transform, which supports the noise insensitivity of the proposed method.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.17 no.1
• /
• pp.90-95
• /
• 2014

The system design and the system performance analysis of 3D imaging laser radar system for the mapping purpose is addressed in this article. For the mapping, a push-bloom scanning method is utilized. The pulsed fiber laser with high pulse energy and high pulse repetition rate is used for the light source of laser radar system. The high sensitive linear mode InGaAs avalanche photo-diode is used for the laser receiver module. The time-of-flight of laser pulse from the laser to the receiver is calculated by using high speed FPGA based signal processing board. To reduce the walk error of laser pulse regardless of the intensity differences between pulses, the time of flight is measured from peak to peak of laser pulses. To get 3D image with a single pixel detector, Risley scanner which stirs the laser beam in an ellipsoidal pattern is used. The system laser energy budget characteristics is modeled using LADAR equation, from which the system performances such as the pulse detection probability, false alarm and etc. are analyzed and predicted. The test results of the system performances are acquired and compared with the predicted system performance. According to test results, all the system requirements are satisfied. The 3D image which was acquired by using the laser radar system is also presented in this article.

• Journal of radiological science and technology
• /
• v.24 no.2
• /
• pp.35-40
• /
• 2001

Performance of SPECT imaging systems which use a rotating gamma camera, are affected by characteristics of the detector-collimator assembly, the data acquisition method, and the filter used in imaging reconstruction. The purpose of this study Is to examine image qualifies of SPECT with different types of low energy collimators. The SPECT imaging system in this study is a digital gamma camera system GCA-901A(Toshiba) and a data processing unit Scintipac-700(Shimadzu). The four types of collimators compared are UHR(ultra high resolution), LEHR(low energy high resolution), LEGP(low energy general purpose), and I-123 PAR(Parallel), with 0.27, 0.66, 1.00, and 2.06 relative sensitivity, respectively. In this case of the same collimators, the spatial resolutions measured in the slice plane showed a slight difference in the FWHM values(mean values of UHR, LEHR, LEGP, and I-123 PAR were 11.3 mm, 13.6 mm, 15.8 mm, and 20.4 mm, respectively) between the center and the circumference of the field of view, in the radial direction, but a large difference in the tangential direction, with lower FWHM values(values of UHR, LEHR, LEGP, and I-123 PAR were 8.4 mm, 8.7 mm, 9.3 mm, and 10.8 mm at 12 cm from the center, respectively). In comparison of SPECT images with the four types of collimators, except for the I-123 PAR collimator, image qualities of UHR, LEHR, and LEGP collimators showed only a slight difference. From the point of for, it is expected that the LEGP collimator would be suitable for SPECT imaging with $^{99m}Tc$.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.21 no.1
• /
• pp.29-33
• /
• 2017

Purpose In this study, we evaluated image by applying normalization factor during 30 days to the PET images. Materials and Methods Normalization factor was acquired during 30 days. We compared with 30 normalization factors. We selected 3 clinical case (PNS study). We applied for normalization factor to PET raw data and evaluated SUV and count (kBq/ml) by drawing ROI to liver and lesion. Results There is no significant difference normalization factor. SUV and count are not different for PET image according to normalization factor. Conclusion We can get a lot of information doing the quality assurance such as performance of sinogram and detector. That's why we need to do quality assurance daily.

• Journal of Radiation Industry
• /
• v.9 no.1
• /
• pp.1-7
• /
• 2015

The trend of x-ray image sensor has been evolved from an amorphous silicon sensor to a crystal silicon sensor. A crystal silicon X-ray sensor, meaning a X-ray CIS (CMOS image sensor), is consisted of three transistors (Trs), i.e., a Reset Transistor, a Source Follower and a Select Transistor, and a photodiode. They are highly sensitive to radiation exposure. As the frequency of exposure to radiation increases, the quality of the imaging device dramatically decreases. The most well known effects of a X-ray CIS due to the radiation damage are increments in the reset voltage and dark currents. In this study, a pixel array of a X-ray CIS was made of $20{\times}20pixels$ and this pixel array was exposed to a high radiation dose. The radiation source was Co-60 and the total radiation dose was increased from 1 to 9 kGy with a step of 1 kGy. We irradiated the small pixel array to get the increments data of the reset voltage and the dark currents. Also, we simulated the radiation effects of the pixel by MCNP (Monte Carlo N-Particle) simulation. From the comparison of actual data and simulation data, the most affected location could be determined and the cause of the increments of the reset voltage and dark current could be found.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.30 no.5
• /
• pp.407-417
• /
• 2019

When a free-falling ballistic missile intercepts a maritime target in a sea clutter environment at high grazing angle, detection performance of the ballistic missile's seeker can be rapidly degraded by the effect of sea clutter. To solve this problem, it is necessary to verify the performance of maritime target detection via simulations based on various scenarios. We accomplish this by applying a two-dimensional cell -averaging constant false alarm rate detector to a two-dimensional radar image, which is generated by merging a sea clutter signal at high grazing angle with a maritime target signal corresponding to the signal-to-clutter ratio. Simulation results using a computer-aided design model and commercial numerical electromagnetic solver in various scenarios show that the performance of maritime target detection significantly depends on the grazing and azimuth angles.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.1
• /
• pp.44-49
• /
• 2021

Recently, due to COVID-19, studies have been popularly worked to apply neural network to mask wearing automatic detection system. For applying neural networks, the 1-stage detection or 2-stage detection methods are used, and if data are not sufficiently collected, the pretrained neural network models are studied by applying fine-tuning techniques. In this paper, the system is consisted of 2-stage detection method that contain MTCNN model for face recognition and ResNet model for mask detection. The mask detector was experimented by applying five ResNet models to improve accuracy and fps in various environments. Training data used 17,217 images that collected using web crawler, and for inference, we used 1,913 images and two one-minute videos respectively. The experiment showed a high accuracy of 96.39% for images and 92.98% for video, and the speed of inference for video was 10.78fps.

• International Journal of Computer Science & Network Security
• /
• v.22 no.11
• /
• pp.308-318
• /
• 2022

Noise is a serious issue. While sending images via electronic communication, Impulse noise, which is created by unsteady voltage, is one of the most common noises in digital communication. During the acquisition process, pictures were collected. It is possible to obtain accurate diagnosis images by removing these noises without affecting the edges and tiny features. The New Average High Noise Density Median Filter. (HNDMF) was proposed in this paper, and it operates in two steps for each pixel. Filter can decide whether the test pixels is degraded by SPN. In the first stage, a detector identifies corrupted pixels, in the second stage, an algorithm replaced by noise free processed pixel, the New average suggested Filter produced for this window. The paper examines the performance of Gaussian Filter (GF), Adaptive Median Filter (AMF), and PHDNF. In this paper the comparison of known image denoising is discussed and a new decision based weighted median filter used to remove impulse noise. Using Mean Square Error (MSE), Peak Signal to Noise Ratio (PSNR), and Structure Similarity Index Method (SSIM) metrics, the paper examines the performance of Gaussian Filter (GF), Adaptive Median Filter (AMF), and PHDNF. A detailed simulation process is performed to ensure the betterment of the presented model on the Mini-MIAS dataset. The obtained experimental values stated that the HNDMF model has reached to a better performance with the maximum picture quality. images affected by various amounts of pretend salt and paper noise, as well as speckle noise, are calculated and provided as experimental results. According to quality metrics, the HNDMF Method produces a superior result than the existing filter method. Accurately detect and replace salt and pepper noise pixel values with mean and median value in images. The proposed method is to improve the median filter with a significant change.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.4
• /
• pp.577-586
• /
• 2022

Detection and classification of steel surface defects are critical for product quality control in the steel industry. However, due to its low accuracy and slow speed, the traditional approach cannot be effectively used in a production line. The current, widely used algorithm (based on deep learning) has an accuracy problem, and there are still rooms for development. This paper proposes a method of steel surface defect detection combining EfficientNetV2 for image classification and YOLOv5 as an object detector. Shorter training time and high accuracy are advantages of this model. Firstly, the image input into EfficientNetV2 model classifies defect classes and predicts probability of having defects. If the probability of having a defect is less than 0.25, the algorithm directly recognizes that the sample has no defects. Otherwise, the samples are further input into YOLOv5 to accomplish the defect detection process on the metal surface. Experiments show that proposed model has good performance on the NEU dataset with an accuracy of 98.3%. Simultaneously, the average training speed is shorter than other models.

• Korean Journal of Optics and Photonics
• /
• v.34 no.3
• /
• pp.106-116
• /
• 2023

In this paper, we designed a coaxial dual camera incorporating two optical systems-one for the visible rays and the other for far-infrared ones-with the aim of capturing images in both wavelength ranges. The far-infrared system, which uses an uncooled detector, has a sensor array of 640×480 pixels. The visible ray system has 1,945×1,097 pixels. The coaxial dual optical system was designed using a hot mirror beam splitter to minimize heat transfer caused by infrared rays in the visible ray optical system. The optimization process revealed that the final version of the dual camera system reached more than 90% of the fusion performance between two separate images from dual systems. Multiple rigorous testing processes confirmed that the coaxial dual camera we designed demonstrates meaningful design efficiency and improved image conformity degree compared to existing dual cameras.