“…First and second images are example of different models of kWh-meter with different size of stand kWh-meter that are successfully extracted by using the flexible size of SE matrix. The other image have similar results to [24]. …”
Section: Stand Kwh-meter Location Detectionsupporting
confidence: 73%
“…This process successfully detected and extracted 145 stand kWh-meters from 155 kWh-meter images (93.55%). The result shows that after using flexible size of SE matrix of morphology operation, the accuracy is better than [24]. The different size of stand kWh-meter that was failed in [24], is successfully detected and extracted.…”
Section: Stand Kwh-meter Location Detectionmentioning
confidence: 68%
“…1 shows the architecture of the proposed system consisting of stand kWh-meter location detection [24], number area segmentation, and numeral recognition process. There are 155 kWh-meter images as the input of the proposed system which taken from PT.…”
Section: Methodsmentioning
confidence: 99%
“…2. In order to minimise the detection error, the detection process in the work of [24] is improved by changing the size of SE matrix of morphology operation, either on erosion or dilation operation, from fixed value to flexible value and giving the iteration from morphology operation to the last step for decreasing size of SE matrix on each iteration. The detection process is iterated until the size ratio of the extracted object is less than the expected ratio.…”
Section: Stand Kwh-meter Location Detectionmentioning
Most of the processes of kilowatt-hour meter (kWh-meter) reading in Indonesia are still in manual process which may lead to some problems, such as time consumption and high possibility of data entry errors. Therefore, this study proposes an automated system to minimise these problems. This system is developed for the image with uneven illumination condition and tilted position of stand kWh-meter due to the unavoidable situation while capturing the kWh-meter image. In this study, the illumination problem is solved by local thresholding and the tilted position of stand kWh-meter is solved by combination of morphology operations and vertical edge detection on the location detection process and vertical-horizontal projections on the segmentation process. Finally, the numeral recognition is performed by support vector machine (SVM) classifier with zonal density feature as a selected input. The results show that the accuracy of proposed system is 93.55% on detection location process, 89.38% on segmentation process, and 78.10% on numeral recognition process.
“…First and second images are example of different models of kWh-meter with different size of stand kWh-meter that are successfully extracted by using the flexible size of SE matrix. The other image have similar results to [24]. …”
Section: Stand Kwh-meter Location Detectionsupporting
confidence: 73%
“…This process successfully detected and extracted 145 stand kWh-meters from 155 kWh-meter images (93.55%). The result shows that after using flexible size of SE matrix of morphology operation, the accuracy is better than [24]. The different size of stand kWh-meter that was failed in [24], is successfully detected and extracted.…”
Section: Stand Kwh-meter Location Detectionmentioning
confidence: 68%
“…1 shows the architecture of the proposed system consisting of stand kWh-meter location detection [24], number area segmentation, and numeral recognition process. There are 155 kWh-meter images as the input of the proposed system which taken from PT.…”
Section: Methodsmentioning
confidence: 99%
“…2. In order to minimise the detection error, the detection process in the work of [24] is improved by changing the size of SE matrix of morphology operation, either on erosion or dilation operation, from fixed value to flexible value and giving the iteration from morphology operation to the last step for decreasing size of SE matrix on each iteration. The detection process is iterated until the size ratio of the extracted object is less than the expected ratio.…”
Section: Stand Kwh-meter Location Detectionmentioning
Most of the processes of kilowatt-hour meter (kWh-meter) reading in Indonesia are still in manual process which may lead to some problems, such as time consumption and high possibility of data entry errors. Therefore, this study proposes an automated system to minimise these problems. This system is developed for the image with uneven illumination condition and tilted position of stand kWh-meter due to the unavoidable situation while capturing the kWh-meter image. In this study, the illumination problem is solved by local thresholding and the tilted position of stand kWh-meter is solved by combination of morphology operations and vertical edge detection on the location detection process and vertical-horizontal projections on the segmentation process. Finally, the numeral recognition is performed by support vector machine (SVM) classifier with zonal density feature as a selected input. The results show that the accuracy of proposed system is 93.55% on detection location process, 89.38% on segmentation process, and 78.10% on numeral recognition process.
“…Adanya pengolahan citra sebenarnya dapat membuat proses pencatatan menjadi lebih efisien dari sisi waktu. Beberapa metode pengolahan citra untuk proses pengambilan citra angka meter tersebut juga sudah dikembangkan [10] [11].…”
Abstrak -Perkembangan teknologi komputer membuat pengolahan citra saat ini banyak dikembangkan untuk dapat membantu manusia di berbagai bidang pekerjaan. Namun, tidak semua bidang pekerjaan dapat dikembangkan dengan pengolahan citra karena tidak mendukung penggunaan komputer sehingga mendorong pengembangan pengolahan citra dengan mikrokontroler atau mikroprosesor khusus. Perkembangan mikrokontroler dan mikroprosesor memungkinkan pengolahan citra saat ini dapat dikembangkan dengan embedded computer atau single board computer (SBC). Penelitian ini bertujuan untuk menguji kemampuan embedded computer dalam mengolah citra dan membandingkan hasilnya dengan komputer pada umumnya (general purpose computer). Pengujian dilakukan dengan mengukur waktu eksekusi dari empat operasi pengolahan citra yang diberikan pada sepuluh ukuran citra. Hasil yang diperoleh pada penelitian ini menunjukkan bahwa optimasi waktu eksekusi embedded computer lebih baik jika dibandingkan dengan general purpose computer dengan waktu eksekusi rata-rata embedded computer adalah 4-5 kali waktu eksekusi general purpose computer dan ukuran citra maksimal yang tidak membebani CPU terlalu besar untuk embedded computer adalah 256x256 piksel dan untuk general purpose computer adalah 400x300 piksel.Kata kunci -waktu eksekusi, bitmap, linux, flip, binerisasi, inversi, mean filter, single board computer (SBC) Abstract -The improvements in computer technology make image processing developed widely to help people in various areas of work. However, some work can't be developed with image processing because it doesn't support the use of computers and it encourages the development of image processing with a microcontroller or special microprocessor. The improvement of microcontrollers and microprocessors features currently allows image processing can be developed with embedded computers or single board computer (SBC). This study aims to test the performance of embedded computers for image processing and then compare the results with the performance of general purpose computer. The result shows that the optimized execution time of embedded computer is better than general purpose computer with the comparison of average execution time of embedded computer is 4-5 times slower than the general purpose computer and the maximal image size that does not make the CPU overload for embedded computer is 256x256 pixel and for general purpose computer is 400x300 pixels.
In the maintenance of Measuring and Limiting Devices (APP) by replacing the kWh Meter on the old kWh meter, as well as P2TL efforts to examine customers who have the potential to commit violations or lack of billing in adjusting electricity rates. In this operation an error / deviation check will be performed on the kWh meter to determine the feasibility of the kWh meter. So far, the measurement of deviation on the kWh Meter 3 Phase is done manually so that it is considered less effective. With this deviation gauge kwh meter 3 phase measurement, meter change officers and P2TL field officers will be able to help in measuring the deviation in the 3 phase kWh meter. This tool reads the power of the kWh meter through the display of the led impulse indicator, then compares it with the power measurement using the metering module at the same time so that the deviation can be detected whether more or less from the meter class, because the measurement standards are based on the meter class. The results obtained from testing on a kWh meter with a grade of 0.5, obtained a deviation of less than 0.5 percent, which means the kWh meter is good because it is still awake in its class according to SPLN No.96 of 1993.Keywords: Deviation, Kwh Meter, Maintaining APP, P2TL
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.