• Journal of Information Processing Systems
• /
• 제16권5호
• /
• pp.1015-1033
• /
• 2020

With promising results and enormous capability, deep learning technology has attracted more and more attention to both theoretical research and applications for a variety of image processing and computer vision tasks. In this paper, we investigate 32 research contributions that apply deep learning techniques to the agriculture domain. Different types of deep neural network architectures in agriculture are surveyed and the current state-of-the-art methods are summarized. This paper ends with a discussion of the advantages and disadvantages of deep learning and future research topics. The survey shows that deep learning-based research has superior performance in terms of accuracy, which is beyond the standard machine learning techniques nowadays.

• 한국멀티미디어학회논문지
• /
• 제22권12호
• /
• pp.1356-1367
• /
• 2019

This paper proposes a novel active stereo matching technique using white auxiliary stripe pattern. The conventional active stereo matching techniques that uses two cameras and an active source such as projector can accurately estimate disparity information even in the areas with low texture compared to the passive ones. However, it is difficult that the conventional active stereo matching techniques using color code patterns acquire these patterns robustly if the object is composed of various colors or is exposed to complex lighting condition. To overcome this problem, the proposed method uses an additional white auxiliary stripe pattern to get and localize the color code patterns robustly. This paper proposes a process based on adaptive thresholding and thinning to obtain the auxiliary pattern accurately. Experimental results show that the proposed method more precisely measures the stepped sample whose depth is known in advance than the conventional method.

• 반도체디스플레이기술학회지
• /
• 제15권4호
• /
• pp.97-102
• /
• 2016

Image stitching is a process of combining two or more images with overlapping area to create a panorama of input images, which is considered as an active research area in computer vision, especially in the field of augmented reality with 360 degree images. Image stitching techniques can be categorized into two general approaches: direct and feature based techniques. Direct techniques compare all the pixel intensities of the images with each other, while feature based approaches aim to determine a relationship between the images through distinct features extracted from the images. This paper proposes a novel image stitching method based on feature pixels with approximated clustering filter. When the features are extracted from input images, we calculate a meaning of the minutiae, and apply an effective feature extraction algorithm to improve the processing time. With the evaluation of the results, the proposed method is corresponding accurate and effective, compared to the previous approaches.

• Journal of Biosystems Engineering
• /
• 제24권3호
• /
• pp.225-234
• /
• 1999

This study was performed as a basic research for establishing an objective quality evaluation method on whole bulbs of garlic. The size of a whole bulb of garlic, the number and the uniformity of complete individual garlics, and the existence of bad individual garlics in the whole bulb of garlic were selected as quality grading factors. Quality discrimination algorithms with machine vision techniques were developed and verified for the four factors based on morphological and physical features of whole bulbs of garlic. Based on the results, the size discrimination by the projected area of a whole bulbs of garlic suggested four grading levels and the algorithm for predicting the number of complete individual garlics based on the peaks on its projected boundary showed ${\pm}$0.78 prediction error. In addition, the uniformity represented by coefficient of variation could be divided into four levels, but the algorithm for discriminating the existence of bad individual garlics in a whole bulb of garlic was not effective.

• 한국멀티미디어학회:학술대회논문집
• /
• 한국멀티미디어학회 2012년도 춘계학술발표대회논문집
• /
• pp.43-45
• /
• 2012

This paper is based on embedding the biometrics techniques on GPU for better computational efficiency and fast matching process using the parallel nature of the GPU processors to compare thousands of images for fingerprint recognition in a fraction of a second. In this paper we worked on GPU (INVIDIA GeForce GTX 260 with compute capability 1.3 and dual core-2-dou processor) for fingerprint matching and found that the efficiency is better than the results with related work already done on CMOS, CPU, ARM9, MATLAB Neural Networks etc which shows the better performance of our system in terms of computational time. The features matching process proposed for fingerprint recognition and the verification procedure is done on 5,000 images which are available online in the databases FVC2000, 2002, 2004 [1].

• ETRI Journal
• /
• 제26권5호
• /
• pp.497-500
• /
• 2004

In this paper, we present a novel idea to integrate a low cost inertial measurement unit (IMU) and Global Positioning System (GPS) for land vehicle localization. By taking advantage of positioning data calculated from an image based on photogrammetry and stereo-vision techniques, errors caused by a GPS outage for land vehicle localization were significantly reduced in the proposed bimodal approach. More specifically, positioning data from the photogrammetric approach are fed back into the Kalman filter to reduce and compensate for IMU errors and improve the performance. Experimental results are presented to show the robustness of the proposed method, which can be used to reduce positioning errors caused by a low cost IMU when a GPS signal is not available in urban areas.

• International Journal of Advanced Culture Technology
• /
• 제8권4호
• /
• pp.119-128
• /
• 2020

Object detection is an important field of computer vision and is applied to applications such as security, autonomous driving, and face recognition. Recently, as the application of artificial intelligence technology including deep learning has been applied in various fields, it has become a more powerful tool that can learn meaningful high-level, deeper features, solving difficult problems that have not been solved. Therefore, deep learning techniques are also being studied in the field of object detection, and algorithms with excellent performance are being introduced. In this paper, a deep learning-based object detection algorithm used to detect multiple objects in an image is investigated, and future development directions are presented.

• 드라이브 ㆍ 컨트롤
• /
• 제19권3호
• /
• pp.16-22
• /
• 2022

Crushers are equipment that crush natural stones, to produce aggregates used at construction sites. As the crusher proceeds, the inner liner becomes worn, causing the size of the aggregate produced to gradually increase. The vision sensor-based aggregate analysis system analyzes the size and distribution of aggregates in production, in real time through image analysis. This study developed an algorithm that can segmentate aggregates in images in real time. using image preprocessing technology combining various filters and morphology techniques, and aggregate region characteristics such as convex hull and concave hull. We applied the developed algorithm to fine aggregate, intermediate aggregate, and thick aggregate images to verify their performance.

• Smart Structures and Systems
• /
• 제32권2호
• /
• pp.111-121
• /
• 2023

This study proposes an automated assembly performance evaluation method for prefabricated steel structures (PSSs) using machine learning methods. Assembly component images were segmented using a modified version of the receptive field pyramid. By factorizing channel modulation and the receptive field exploration layers of the convolution pyramid, highly accurate segmentation results were obtained. After completing segmentation, the positions of the bolt holes were calculated using various image processing techniques, such as fuzzy-based edge detection, Hough's line detection, and image perspective transformation. By calculating the distance ratio between bolt holes, the assembly performance of the PSS was estimated using the k-nearest neighbors (kNN) algorithm. The effectiveness of the proposed framework was validated using a 3D PSS printing model and a field test. The results indicated that this approach could recognize assembly components with an intersection over union (IoU) of 95% and evaluate assembly performance with an error of less than 5%.

• 드라이브 ㆍ 컨트롤
• /
• 제19권1호
• /
• pp.1-7
• /
• 2022

Recently, a re-painting robot was developed to semi-automatically conduct blasting work in bridge spaces to improve work productivity and worker safety. In this study, a vision sensor-based monitoring module was developed to automatically move the re-painting robot along the path. The monitoring module provides direction information to the robot by analyzing the boundary between the painting surface and the metal surface. To stably measure images in unstable environments, various techniques for improving image visibility were applied in this study. Then, the driving performance was verified in a similar environment.