• Journal of the Semiconductor & Display Technology
• /
• v.15 no.3
• /
• pp.51-56
• /
• 2016

This paper proposes a new image enhancement method for an infrared thermal image. The proposed method uses both Laplacian and Prewitt edge detectors. Without a visible light, it uses an infrared image for the edge detection. The method subtracts contour images from the infrared thermal image. It results black contours of objects in the infrared thermal image. That makes the objects in the infrared thermal image distinguished clearly. The proposed method is implemented using C language in an embedded Linux system for a high-speed real-time image processing. Experiments were conducted by using various infrared thermal images. The results show that the proposed method is successful for image enhancement of an infrared thermal image.

• Journal of the Semiconductor & Display Technology
• /
• v.13 no.4
• /
• pp.49-54
• /
• 2014

This paper proposes a new color segmentation method of vehicle license plates in the RGB color space. Firstly, the proposed method shifts the histogram of an input image rightwards and then stretches the image of the histogram slide. Secondly, the method separates each of the three RGB color components and performs the adaptive threshold processing with the three components, respectively. Finally, it combines the three components under the condition of making up a segment color and removes noises with the morphological processing. The proposed method is implemented using C language in an embedded Linux system for a high-speed real-time image processing. Experiments were conducted by using real vehicle images. The results show that the proposed algorithm is successful for most vehicle images. However, the method fails in some vehicles when the body and the license plate have the same color.

• Journal of the Semiconductor & Display Technology
• /
• v.20 no.3
• /
• pp.131-135
• /
• 2021

Numbers of studies related to optimization of design of organic light emitting diodes(OLED) through machine learning are increasing. We propose the generative method of the image to assess the performance of the device combining with machine learning technique. Principle parameter regarding dark spot growth mechanism of the OLED can be the key factor to determine the long-time performance. Captured images from actual device and randomly generated images at specific time and initial pinhole state are fed into the deep neural network system. The simulation reinforced by the machine learning technique can predict the device parameters accurately and faster. Similarly, the inverse design using multiple layer perceptron(MLP) system can infer the initial degradation factors at manufacturing with given device parameter to feedback the design of manufacturing process.

• Journal of the Semiconductor & Display Technology
• /
• v.20 no.3
• /
• pp.45-50
• /
• 2021

The adjustment of lens magnification can make the degree of precision in pattern measurement be improved, but several problems such as high cost, smaller field of view and stage error accumulation are followed. In this paper, a method for precisely measuring patterns is proposed based on gradient transition vector, in order to solve these problems. The performance of our method is evaluated using pattern images with several directions. Also, it is compared with previous methods based on edge and gray-level moment. It is judged that the proposed method outperforms consistent pattern width results, and so could be applied to automation processes for measurement and inspection of precise and complexed patterns in IT, BT industry products.

• Journal of the Semiconductor & Display Technology
• /
• v.19 no.4
• /
• pp.34-38
• /
• 2020

In this paper, we implement a Dashcam system capable of shooting 360 degrees using a Raspberry Pi, shock sensors, distance sensors, and rotating camera with a servo motor. If there is an object approaching the vehicle by the distance sensor, the camera rotates to take a video. In the event of an external shock, videos and images are stored in the server to analyze the cause of the vehicle's accident and prevent the user from forging or tampering with videos or images. We also implement functions that transmit the message with the location and the intensity of the impact when the accident occurs and send the vehicle information to an insurance authority with by linking the system with a smart device. It is advantage that the authority analyzes the transmitted message and provides the accident handling information giving the user's safety and convenience.

• Journal of Semiconductor Engineering
• /
• v.2 no.1
• /
• pp.125-129
• /
• 2021

As the interest of deep learning, techniques to control the color of images in image processing field are evolving together. However, there is no clear standard for color, and it is not easy to find a way to represent only the color itself like the color-palette. In this paper, we propose a novel color palette extraction system by chroma fine-tuning with reinforcement learning. It helps to recognize the color combination to represent an input image. First, we use RGBY images to create feature maps by transferring the backbone network with well-trained model-weight which is verified at super resolution convolutional neural networks. Second, feature maps are trained to 3 fully connected layers for the color-palette generation with a generative adversarial network (GAN). Third, we use the reinforcement learning method which only changes chroma information of the GAN-output by slightly moving each Y component of YCbCr color gamut of pixel values up and down. The proposed method outperforms existing color palette extraction methods as given the accuracy of 0.9140.

• Journal of the Semiconductor & Display Technology
• /
• v.21 no.4
• /
• pp.156-162
• /
• 2022

Image compression plays an important role in encoding and improving various forms of images in the digital era. Recent researches have focused on the principle of deep learning as one of the most exciting machine learning methods to show that it is good scheme to analyze, classify and compress images. Various neural networks are able to adapt for image compressions, such as deep neural networks, artificial neural networks, recurrent neural networks and convolution neural networks. In this review paper, we discussed how to apply the rule of deep learning to obtain better image compression with high accuracy, low loss-ness and high visibility of the image. For those results in performance, deep learning methods are required on justified manner with distinct analysis.

• Journal of the Semiconductor & Display Technology
• /
• v.21 no.4
• /
• pp.77-80
• /
• 2022

Automatic Vision Inspection (AVI) systems automatically detect defect features and measure their sizes via camera vision. It has been populated because of the accuracy and consistency in terms of QC (Quality Control) of inspection processes. Also, it is important to predict the performance of an AVI to meet customer's specification in advance. AVI are usually suffered from false negative and positives. It can be overcome by providing extra information such as 3D depth information. Stereo vision processing has been popular for depth extraction of the 3D images from 2D images. However, stereo vision methods usually take long time to process. In this paper, retro optical system using reflectors is proposed and experimented to overcome the problem. The optical system extracts the depth without special SW processes. The vision sensor and optical components such as illumination and depth detecting module are integrated as a unit. The depth information can be extracted on real-time basis and utilized and can improve the performance of an AVI system.

• Journal of the Semiconductor & Display Technology
• /
• v.22 no.2
• /
• pp.40-44
• /
• 2023

The defect detection based on deep learning requires minimal loss and high accuracy to pinpoint product defects. In this paper, we confirm the loss rate of deep learning training based on disc-shaped artificial scaffold images. It is intended to compare the performance of Cross-Entropy functions used in object detection algorithms. The model was constructed using normal, defective artificial scaffold images and category cross entropy and sparse category cross entropy. The data was repeatedly learned five times using each loss function. The average loss rate, average accuracy, final loss rate, and final accuracy according to the loss function were confirmed.

• Journal of the Semiconductor & Display Technology
• /
• v.23 no.1
• /
• pp.56-60
• /
• 2024

In this paper, we propose a SIFT feature-based dynamic stitching algorithm for image calibration and correction of a 360-degree surround view system. The existing surround view system requires a lot of processing time and money because in the process of image calibration and correction. The traditional marker patterns are placed around the vehicle and correction is performed manually. Therefore, in this study, images captured with four fisheye cameras mounted on the surround view system were distorted and then matched with the same feature points in adjacent images through SIFT-based feature point extraction to enable image stitching without a fixed marker pattern.