• 한국멀티미디어학회논문지
• /
• 제17권12호
• /
• pp.1418-1429
• /
• 2014

In this paper we suggest a new two-wheelers detection algorithm using local cell features. The first, we propose new feature vector matrix extraction algorithm using the correlation two cells based on local cell histogram and shifting from the result of histogram of oriented gradients(HOG). The second, we applied new weighting values which are calculated by the modified histogram intersection showing the similarity of two cells. This paper applied the Adaboost algorithm to make a strong classification from weak classification. In this experiment, we can get the result that the detection rate of the proposed method is higher than that of the traditional method.

• 말소리와 음성과학
• /
• 제4권2호
• /
• pp.79-86
• /
• 2012

In this paper, we propose a novel approach to improve histogram equalization for speaker identification. Our method collects all speech features of UBM training data to make a reference distribution. The ranks of the feature vectors are calculated in the sorted list of the collection of the UBM training data and the test data. We use the ranks to perform order-based histogram equalization. The proposed method improves the accuracy of the speaker recognition system with short utterances. We use four kinds of speech databases to evaluate the proposed speaker recognition system and compare the system with cepstral mean normalization (CMN), mean and variance normalization (MVN), and histogram equalization (HEQ). Our system reduced the relative error rate by 33.3% from the baseline system.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2004년도 심포지엄 논문집 정보 및 제어부문
• /
• pp.111-113
• /
• 2004

Histogram Equalization is the most popular algorithm for contrast enhancement due to its effectiveness and simplicity. In this paper, We propose the advanced contrast enhancement method using genetic algorithm. We propose a novel objective criterion for enhancement, and attempt finding the best image according to the respective criterion. Due to the high complexity of the enhancement criterion proposed, we employ a Genetic Algorithm. We compared our method with other enhancement techniques, like Global Histogram Equalization and Partially Overlapped Sub-Block Histogram Equalization(POSHE).

• 한국멀티미디어학회논문지
• /
• 제19권8호
• /
• pp.1288-1296
• /
• 2016

The reliable detection of the limited defect in TFT-LCD images is difficult due to the small intensity difference with the background. However, the proposed detection method reliably detects the limited defect by enhancing the TFT-LCD image based on the cumulative histogram and then detecting the defect through the mean and standard deviation of the enhanced image. Notably, an image enhancement using a cumulative histogram increases the intensity contrast between the background and the limited defect, which then allows defects to be detected by using the mean and standard deviation of the enhanced image. Furthermore, through the comparison with the histogram equalization, we confirm that the proposed algorithm suppresses the emphasis of the noise. Experimental comparative results using real TFT-LCD images and pseudo images show that the proposed method detects the limited defect more reliably than conventional methods.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권4호
• /
• pp.2113-2128
• /
• 2019

Image thresholding techniques introducing spatial information are widely used image segmentation. Some methods are used to calculate the optimal threshold by building a specific histogram with different parameters, such as gray value of pixel, average gray value and gradient-magnitude, etc. However, these methods still have some limitations. In this paper, an entropic thresholding method based on Gabor histogram (a new 2D histogram constructed by using Gabor filter) is applied to image segmentation, which can distinguish foreground/background, edge and noise of image effectively. Comparing with some methods, including 2D-KSW, GLSC-KSW, 2D-D-KSW and GLGM-KSW, the proposed method, tested on 10 realistic images for segmentation, presents a higher effectiveness and robustness.

• International Journal of Internet, Broadcasting and Communication
• /
• 제13권2호
• /
• pp.156-165
• /
• 2021

In this paper, we propose an histogram weighted centroid based Siamese region proposal network for object tracking. The original Siamese region proposal network uses two identical artificial neural networks which take two different images as the inputs and decide whether the same object exist in both input images based on a similarity measure. However, as the Siamese network is pre-trained offline, it experiences many difficulties in the adaptation to various online environments. Therefore, in this paper we propose to incorporate the histogram weighted centroid feature into the Siamese network method to enhance the accuracy of the object tracking. The proposed method uses both the histogram information and the weighted centroid location of the top 10 color regions to decide which of the proposed region should become the next predicted object region.

• 전자공학회논문지
• /
• 제51권11호
• /
• pp.155-164
• /
• 2014

지역 히스토그램 평활화 방법은 입력 영상의 국부적은 밝기 특성을 부각시키기 위한 가장 널리 사용되는 방법들 중 하나이다. 그러나 지역 히스토그램 평활화 기반의 방법들은 몇 가지의 문제점들을 발생시킨다. 먼저, 국부적인 특성들을 과도하게 부각시켜 의도하지 않는 결함들을 발생시킨다. 두 번째, 국부 특성들의 향상이 전역 콘트라스트 향상을 증대시키지는 않는다는 점이다. 이러한 문제들을 해결하기 위해, 우리는 광원 정보를 이용한 지역 히스토그램 평활화 방법을 제안한다. 먼저, 광원 정보를 추정하기 위하여 제안하는 방법은 입력 영상의 다운 샘플링과 업 샘플링 과정을 통하여 획득된 블러 영상과 원 영상을 융합한다. 그 후, 지역 히스토그램 평활화 방법에서 추정한 변환 함수를 광원 정보를 이용하여 적응적으로 조절한다. 그 결과 기존 방법에서 발생할 수 있는 결함을 억제시키면서, 전역 콘트라스트와 국부 콘트라스트를 동시에 향상시킬 수 있다. 실험 결과들은 제안하는 방법이 기존 방법에 비해 수치적인 면과 시각적인 면에서 뛰어난 결과를 보임을 확인할 수 있다.

• Transactions on Electrical and Electronic Materials
• /
• 제13권1호
• /
• pp.10-15
• /
• 2012

In conventional, skin detection methods using for skin color definitions is based on prior knowledge. By experimentation, the threshold value for dividing the background from the skin region is determined subjectively. A drawback of such techniques is that their performance is dependent on a threshold value which is estimated from repeated experiments. To overcome this, the present paper introduces a skin region detection method. This method uses a histogram approximation based on the mean shift algorithm. This proposed method applies the mean shift procedure to a histogram of a skin map of the input image. It is generated by comparing with the standard skin colors in the $C_bC_r$ color space. It divides the background from the skin region by selecting the maximum value according to the brightness level. As the histogram has the form of a discontinuous function. It is accumulated according to the brightness values of the pixels. It is then, approximated by a Gaussian mixture model (GMM) using the Bezier curve technique. Thus, the proposed method detects the skin region using the mean shift procedure to determine a maximum value. Rather than using a manually selected threshold value, as in existing techniques this becomes the dividing point. Experiments confirm that the new procedure effectively detects the skin region.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2006년 학술대회 논문집 정보 및 제어부문
• /
• pp.431-433
• /
• 2006

The goal of this paper is improvement of vision inspection accuracy by using histogram specification operation. The histogram is composed of horizontal axis of image intensity value and vertical axis of pixel number in image. In appearance vision inspection, the histogram of reference image and input image are different because of minutely lighting distinction. The minutely lighting distinction is main reason of vision inspection error in many cases. Therefore we made an effort for elevation of vision inspection accuracy by making the identical histogram of reference image and input image. As a result of this area separation histogram specification algorithm, we could increase the exactness of vision inspection and prevent system error from physical and spirit condition of human. Also this system has been developed only using PC, CCD Camera and Visual C++ for universal workplace.

• 방송공학회논문지
• /
• 제19권1호
• /
• pp.96-108
• /
• 2014

기존 히스토그램 평활화 방법을 사용하여 영상의 명암대비를 증가시킬 경우 과도한 밝기 변화로 인한 과포화 현상(over-enhancement), 계조현상(false contouring) 및 영상의 세부 정보가 없어지는 등의 왜곡이 발생한다. 특히 밝기 분포가 특정한 밝기 레벨에 밀집되어 있는 경우 이러한 왜곡이 두드러지게 나타나게 된다. 이러한 문제를 해결하기 위하여 임계치를 이용한 히스토그램 클리핑을 통해 입력 히스토그램을 변형하는 개선된 평활화 방법들이 제시되었지만, 입력영상의 히스토그램 특성을 고려하지 않고 전체 히스토그램에 대해 동일한 임계치를 적용하기 때문에 명암대비 향상효과가 감소하고, 입력 영상의 특성을 유지하지 못해 부자연스러운 영상이 얻어지기도 한다. 본 논문에서는 기존 방식에서 발생하는 문제를 해결하기 위하여 입력영상의 히스토그램의 빈도수에 따른 차별적 압축방법을 적용하여 과도한 밝기 변화가 발생하는 문제를 억제하면서도 입력영상의 특성을 유지하는 새로운 평활화 방식을 제 안한다. 또한 입력영상의 특성에 따라 압축률의 강도를 제어하여 보다 효과적으로 명암대비 향상을 수행하는 방법을 제시한다.