Shadow Elimination in Soccer Game Video using Background Subtraction and Sobel Operators

Object detection in real time is considered as a challenging problem. However, it is very important in a wide range of applications, especially in field of multimedia. The players and ball are the most important objects in soccer game videos and detecting them is a challenging task because of many difficulties, such as shadow and illumination, ball size, ball occluded by players or merged with lines, and similar appearance of players. To overcome these problems, we present a new system to detect the players and ball in real-time by using background subtraction and Sobel detection. The results were more accurate and approximately two times faster than those using only background subtraction.

Mendeteksi Tepi Citra Penyakit Hemokromatosis Dengan Menggunakan Metode Log (Laplacian Of Gaussian)

Hemochromatosis is a genetic or hereditary disease. Abnormalities of iron metabolism characterized by excessive deposition of iron in the tissues. Derivative conditions that cause the body to absorb too much iron from the food eaten. Excess iron is stored in organs such as the liver, heart and pancreas. Excess iron can cause toxicity to these organs, and is life threatening because it can cause diseases such as cancer, cardiac arrhythmias, and cirrhosis. LOG method (Laplacian of Gaussian) is a second-order edge detection operator or has a derivative filter whose function can detect areas that have rapid changes (rapit change) such as edges (edges) in the image. But this laplacian is very sensitive or low to the presence of noise. For that, the image will be smoothed first by using Gaussian. Thus a new derivative function is known, namely LoG or Laplacian of Gaussian. Many methods are used in solving edge detection problems, including Prewitt Operators, Sobel Operators, Canny Operators, but among all these methods, the Laplacian of Gaussian method is the method most often used in detecting edges. For this reason, it is hoped that the LoG (Laplacian of Gaussian) method can help to detect hemochromatosis. Can help how severe the disease has developed, and what symptoms can later be caused, so that it can help in the healing process.

Analisis Perbandingan Perbaikan Kualitas Citra Pada Motif Batik Dengan Konsep Deteksi Tepi Robert, Sobel, Canny Menggunakan Metode Morfologi

Image results from quality edge detection are not optimal. From these problems a method is needed to improve the image quality of edge detection. The method used is Dilation Morphology on the results of edge detection of batik patterns. The results of testing the improved image quality of edge detection 10 batik patterns using Dilation Morphology show that Canny operators are able to produce very high accuracy from operators Robert and Sobel, with the percentage of Canny operators is 80%. While Robert operators with a percentage of 40% and Sobel operators 60%. The application of Dilation Morphology to operators Robert, Sobel and Canny can improve image quality of edge detection and improve accuracy in batik patterns.

FPGA IMPLEMENTATION OF HIGH SPEED AND LOW POWER ARCHITECTURES FOR IMAGE SEGMENTATION USING SOBEL OPERATORS

In this paper we present two architectures for image segmentation using Sobel Operators. The first architecture is designed for optimum speed whereas the second one is designed for low power. To improve the speed of operation and to reduce the memory access two identical processing units operate parallelly. The first architecture is able to segment up to 800 images each of 640 × 480 pixels in one second at 500 MHz clock frequency consuming 27.31 mW dynamic power. The second architecture is able to segment up to 488 images each of 640 × 480 pixels in one second at 300 MHz clock frequency consuming 13.6 mW dynamic power.

Research on the Lane Detection Algorithm Based on Zoning Hough Transformation

In order to improve the adaptability of the lane detection algorithm under complex conditions such as damaged lane lines, covered shadow, insufficient light, the rainy day etc. Lane detection algorithm based on Zoning Hough Transform is proposed in this paper. The road images are processed by the improved ±45° Sobel operators and the two-dimension Otsu algorithm. To eliminate the interference of ambient noise, highlight the dominant position of the lane, the Zoning Hough Transform is used, which can obtain the parameters and identify the lane accurately. The experiment results show the lane detection method can extract the lane marking parameters accurately even for which are badly broken, and covered by shadow or rainwater completely, and the algorithm has good robustness.