Retinoblastoma is cancer on the retina that often occurs in infants and children that can cause blindness and even death. The cause of retinoblastoma is due to the mutation of the RB1 gene which keeps retinal cells reproducing until the tumor grows on the retina. In general, to identify a retinoblastoma the doctor uses an ophthalmoscope that shines brightly through the pupil to examine the back of the eye and see the presence of white or yellowish white tumor lesions in the eye. In addition, the examination is also done by analyzing the retinal fundus image from the fundus camera. The image of the fundus camera is re-analyzed by a doctor or expert to determine whether or not there is retinoblastoma. Therefore, a system is needed to help the expert to diagnose retinoblastoma. The method used in this study is the extreme learning machine. Retinal fundus images are used as input images to identify retinoblastoma. Before being identified, pre-processing of the image is carried out, it consists of scaling, green channel, contrast stretching, thresholding, and feature extraction using the zoning method. From this study, it was concluded that the proposed method had the ability to identify retinoblastoma with an accuracy of 92%.
1 June 2019 in Journal of Physics: Conference Series
Compression purpose to reduce the redundancy data as small as possible and speed up the data transmission process. To solve the size problem in saving data and transmission process, we use Run Length Encoding and Fibonacci Code algorithm to do compression process. Run Length Encoding and Fibonacci Code algorithm is a type of lossless data compression used in this research, which performance will be measured by comparison parameters of the Compression Ratio (CR), Redundancy (RD), Space Saving (SS) and Compression Time. The compression process is only done on image files with Bitmap format (*.bmp) and encode using Run Length Encoding or Fibonacci Code, then perform the compression process. The final result of the compression is file with extension *.rle or *.fib which contains compressed information that can be decompressed back. The output of the decompression result is an original image file that is stored with *.bmp extension. Fibonacci algorithm will give a better compressed size on image color, while in a grayscale image Run Length Encoding will give a better compressed size. Based on the results of research at two different types of images, each algorithm has its own advantages. Fibonacci Code algorithm is better for color image compression while Run-Length algorithm Encoding is better for grayscale image compression.
15 March 2018 in Journal of Physics: Conference Series
Security system is one of the common problems to protect an environment such as personal house or a warehouse. There are numerous methods and technologies that can be used as part of a security system. In this paper, we present a security system that offers a better efficiency. The purpose of this study is to build a system that can monitor home security at any time in particular fire and theft. Through sensors, the system will be able to provide warning information of hazard conditions via LCD monitor, sound, and alarm. This information will be sent automatically to the home owner's smartphone as well as to the corresponding to the security agency. Thus the prevention of theft and fire hazards can be immediately anticipated by the police and firefighters. The system will also notify the position of the coordinates of the location of the building (the house) by a link to the Google map in order to make it easier to get the location quickly.
1 June 2019 in Journal of Physics: Conference Series
One of the ways to publish scientific paper is to present it in a conference. Before being presented in the conference, the registered scientific paper has to be grouped into categories. Because there are so many scientific papers that have been registered, it is an urgent need to establish a system that can automatically identify the topic of scientific papers and classify them according to recognized topics. One method that can be used to organize documents according to their categories is classification. In this study, Support Vector Machine algorithm is used as a method to classify documents into 5 categories of computer science. System test is done by taking 150 documents as data sample and 50 documents as data testing. This study produced a classification system of scientific documents and showed the method used was able to classify documents with 90% accuracy results.
1 June 2019 in Journal of Physics: Conference Series
Rainfall is a very important factor in agriculture and development planning. This study aims to predict rainfall and rainfall properties in Medan using a multilayer neural network with Backpropagation and Learning Vector Quantization (LVQ) algorithms. Rainfall data used for training is rainfall data for the last 30 years. The parameters used in this study consisted of two inputs namely the month and the volume of rainfall. The training process produces the best architecture with 3 layer three for Backpropagation and two layers for Learning Vector Quantization with learning rate 0.5. From the prediction results obtained Backpropagation algorithm more accurate in predicting rainfall data last 30 years compared with LVQ with an average difference of 21.99%. Backpropagation and LVQ algorithms have better accuracy in the dry season, with accuracy for Backpropagation algorithm between 75 - 99% and LVQ algorithm of 60 - 82%. For both algorithms, the influence of El-Nino and La-Nina phenomena is not so significant.
1 June 2019 in Journal of Physics: Conference Series
On sending images via the internet, such as email, social media, and other are very vulnerable to technological crimes such as tapping, or modification. Without security guarantees, others can easily get messages sent over the internet. In this research, combined cryptographic algorithms and steganography. Cryptographic methods can ensure the security of data information by convering it into random codes, however making the data can’t be read and understood by others. One of the cryptographic methods used by the author is the Myszkowski Transposition Algorithm. But the use of such encryption method does not always guarantee data security because the data is messy with other words can be regarded as unusual and suspicious. This algorithm also has the disadvantage that as complicated as any we do transposition or permutation on the characters in plaintext, we just do a randomization of the order of plaintext but we don’t change it. So it must be combined with steganography technique that is using Modified Least Significant Bit (MLSB), so the message is not visible inserted visible. Therefore, if the information is on the wrong side will cause losses. The result of this study indicate that the Myszkowski Transposition algorithm and Green Channel Modified Least Significant Bit (MLSB) on PNG image coverage, can restore the image as before. So it can be concluded that the Myszkowski Transposition algorithm and Modified Least Significant Bit (MLSB) is capable of securing the message. Therefore, it can be concluded that the bigger the image size will be the longer the process.