Showing papers by "De Rosal Ignatius Moses Setiadi published in 2017"

Tomatoes classification using K-NN based on GLCM and HSV color space

Abstract: Tomatoes at any given time has different maturity levels, therefore it is necessary to recognize the appropriate pattern to determine the level of maturity. One of the recognition patterns of tomatoes image is to use texture and color analysis. Texture analysis can be processed using the Gray Level Co-occurrence Matrix (GLCM) method. GLCM is chosen because it has a high degree of recognition based on the value of contrast, correlation, homogeneity, and energy. Furthermore, for color analysis one of the methods that can be used is Hue, Saturation, Value (HSV). By utilizing HSV, an object with a certain color can be detected and reduce the influence of the intensity of light from outside. The results of GLCM and HSV calculations can be classified to determined the maturity level of tomatoes by using K-Nearest Neighbour (K-NN) one of basic and simple classification method which utilizes the distance (k) as a comparison of the similarity level of the image. From the research we have done, using 100 data sets, consisting of 75 training data and 25 testing data that yields the highest accuracy rate of 100% with p value on GLCM is 9 and the membership value (k) in K-NN is 3. According to the experimental results, we can conclude that our proposed method can achieve the highest accuracy.

Secure Image Steganography Algorithm Based on DCT with OTP Encryption

Abstract: Rapid development of Internet makes transactions message even easier and faster. The main problem in the transactions message is security, especially if the message is private and secret. To secure these messages is usually done with steganography or cryptography. Steganography is a way to hide messages into other digital content such as images, video or audio so it does not seem nondescript from the outside. While cryptography is a technique to encrypt messages so that messages can not be read directly. In this paper have proposed combination of steganography using discrete cosine transform (DCT) and cryptography using the one-time pad or vernam cipher implemented on a digital image. The measurement method used to determine the quality of stego image is the peak signal to noise ratio (PSNR) and ormalize cross Correlation (NCC) to measure the quality of the extraction of the decrypted message. Of steganography and encryption methods proposed obtained satisfactory results with PSNR and NCC high and resistant to JPEG compression and median filter. Keywords —Image Steganography, Discrete Cosine Transform (DCT), One Time Pad, Vernam, Chiper, Image Cryptography

CBIR for classification of cow types using GLCM and color features extraction

Abstract: Cow is one of the animals that have many benefits for humans. There are various types of cows based on benefits such as dairy cows, beef cattle, worker cattle, and others. Cattle breeding should be tailored to the needs of the public. Less knowledge about different types of cattle can reduce the benefits of farmed cattle. Content Based Image Retrieval (CBIR) can be applied to help the problem of distinguishing or knowing the type of cow. The first step of the method proposed in this research is preprocessing by changing the background color, resizing and conversion of color space. Color feature extraction calculates the average and standard deviation of the color intensity of each color component. Next extract the texture feature using Gray Level Cooccurrence Matrix (GLCM) to look for contrast, energy, correlation, homogeneity and entropy at each angle 0°, 45°, 90° and 135° with a mean of 1 averaged. Six color features and five texture features are used as attributes to perform calculations with Euclidean Distance, so it can be known the similarity between images. Cattle types used include Limousin, Simental, Brangus, Peranakan Ongole (PO), and Frisien Holstein (FH). With 100 training images and 20 test images. To measure the accuracy of the proposed CBIR is used Confusion Matrix. Based on the measurement results obtained accuracy of 95% while the precision and recall obtained 100%.

Hiding and securing message on edge areas of image using LSB steganography and OTP encryption

Abstract: Imperceptibility and security are the two most important things in StegoCrypt techniques In this paper, we have proposed the technique of inserting messages to the edge of the image using the Least Significant Bit (LSB) algorithm The image edge area is selected because human vision can't detect small changes in this area To provide double protection of messages, applied encryption using one-time pad (OTP) algorithm To strengthen the encryption result, message encryption process is not done directly but by converting it into binary message first This way can result in more verbalized message encryption, thereby minimizing the chance of messages being deciphered by unauthorized parties From the experimental results of this paper obtained a better imperceptibility value, where the value is measured by PSNR and MSE Histogram of original image and stego image also shows great similarity identical Meanwhile, message extraction can also be done perfectly with the value of Correlation Coefficient (CC) is 1

A Good Performance OTP Encryption Image based on DCT-DWT Steganography

Abstract: The security aspect is very important in data transmission. One way to secure data is with steganography and cryptography. Surely research on this should continue to be developed to improve security. In this paper, we proposed a combination of steganographic and cryptographic algorithms for double protection during data transmission. The selected steganographic algorithm is the use of a combination of DCT and DWT domain transformations. Because the Imperceptibility aspect is a very important aspect of steganographic techniques, this aspect needs to be greatly improved. In the proposed method of DCT transformation first, proceed with DWT transformation. From the experimental results obtained better imperceptibility quality, compared with existing methods. To add OTP message security applied algorithm to encrypt the message image, before it is inserted. This is evidenced by experiments conducted on 20 grayscale images measuring 512x512 with performance tests using MSE, PSNR, and NC. Experimental results prove that DCT-DWT-OTP generates PNSR more than 50 dB, and NC of all images is 1.

Digital image signature using triple protection cryptosystem (RSA, Vigenere, and MD5)

Abstract: Authentication of digital media has been done with the various scheme, one of them is a digital signature The main reason the technique of digital signature appears on the concerns of people when sending important documents and raises the thought that the file submitted is not changed when it is transmitted and the content is different when received by the recipient of the document The digital signature attached by the sender to the document can be used as a tool to ensure that the submitted document is authentic or not manipulated So research on digital signatures should be developed to improve its ability to provide security and prove the authenticity of the image This paper proposes a combination of three algorithms to the created digital signature, namely: Rivest — Shamir — Adleman (RSA), Vigenere Cipher and Message Digest 5 (MD 5) The proposed method also tested with various attacks to measure the reliability of digital signatures Various attacks which applied such as blurring, salt, and pepper, Gaussian filters Based on the attacking result, the smallest change occurred in the blurring attack has a very good PSNR is 867532 dB The experimental result proves the little change of image and filename can affect the validation result Then, it can be concluded that the proposed method suitable for image authentication

A performance analysis StegoCrypt algorithm based on LSB-AES 128 bit in various image size

Abstract: Many people have used the internet in daily communication. The rise of internet data theft makes the security of data transmission becomes very important. Cryptography and Steganography is a technique that is widely used to secure data so that data can not be accessed by unauthorized persons. Cryptography is a technique of message randomization, while steganography is a data-hiding technique. Combination Stego-Crypto Least Significant Bit (LSB) — Advanced Encryption Standard (AES) is widely proposed in previous studies because of its reliability in providing multiple protection. The secret message is encrypted with the AES method before it is embedded into the image using LSB. In this study, analyzing the performance of a combination of LSB-AES algorithms on various cover image sizes. The cover image used is RGB color images. To measure imperceptibility performance is used Peak Signal to Noise Ratio (PSNR), Mean Square Error (MSE), and Histogram. The measurement is done by comparing the original image with the resulting image. From the test results, we get a very good value PSNR with an average value above 70dB. The results of encryption and decryption can work perfectly. To test cryptographic performance is measured by time and entropy. The results of the test show good cryptographic quality.

An imperceptible LSB image hiding on edge region using DES encryption

Abstract: Sending image in a plain form has a high risk for the data being vulnerable to steal. Thus, due to reducing the risk, data security techniques should be implemented. Data security techniques can be done using Cryptography and Steganography. Based on the literature, the combination of cryptography and steganography can enhance results. Cryptography as one of science in information security, which used in the process of securing image. One of the cryptographic methods that have been used to secure the data is Data Encryption Standard (DES). This algorithm often implemented for the encryption process in an ATM machine and SIM card operator worldwide. Our proposed method firstly encrypts the image message using DES encryption. This process produces visible data, which is caused suspicion by others. In this paper, to conceal the encrypted image, we used steganography method. Our experiment chooses Least Significant Bit (LSB) based on canny edge detection. LSB algorithm was chosen because it is simple to be implemented and has good imperceptible value. Then, canny edge detection used to determines the coordinates of the object's edges of the cover image. We utilize the Canny coordinate result as the position to embed the image message because of the results more accurate and it can improve imperceptibility. Our experimental results were measured using Mean Square Error (MSE) and Peak Signal to Noise Ratio (PSNR). From the measurement, results obtained the best MSE value is 0.0038557 and PSNR are 72.2698 dB.

Hybrid method using 3-DES, DWT and LSB for secure image steganography algorithm

Abstract: Public networks such as the Internet are becoming more sophisticated, faster, and cheaper, so that more and more used for information exchange. This may increase the likelihood of confidential information from being stolen and exploited by unauthorized persons. This study proposed a combination of two Steganography domains coupled with Cryptography which aimed to make confidential information more secure and inaccessible to unauthorized persons. Messages are encrypted using the 3-DES method. On the other side of the cover image is decomposed into four subbands by using DWT. LH, HL, and HH subbands are chosen to embed encrypted message using LSB method. The last step, done Inverse DWT (IDWT) to get the stego image reconstruction. From the proposed method is then measured its quality with PSNR and MSE. As for message encryption results are measured using entropy. From the experiment results obtained PSNR results with a value of 55.30 dB for image messages size 64 ∗ 64 and 49.23 dB for messages size 128 ∗ 128. The extraction process can also be done perfectly with NC 1 and the average of Entropy of encrypted messages are 7.95754 for 64∗64 and 7.98904 for 128∗128.

Fast and efficient image watermarking algorithm using discrete tchebichef transform

Abstract: Discrete Tchebichef Transform (DTT) is a transformation that has been widely applied in image processing, but still not widely applied in image watermarking. Based on the previous research results, DTT has advantages to reduce computational complexity, so that computation can be done faster and efficiently than the Discrete Cosine Transform (DCT). DTT also have similar properties with DCT, such as separability, symmetry, and orthogonality. DTT has also been shown to have higher energy compactness than the DCT, thus making DTT potential to be implemented in the image watermarking algorithm. In this paper, DTT will be introduced and applied in the techniques of image watermarking. Based on the results of this research, there is a very significant difference in the time required for computation, where DTT is proven to be very efficient to reduce algorithm complexity so that the time spent embedding and extracting copyright is about five times faster than DCT. Nevertheless, the watermarked image quality is also well preserved, as is the robust to various attacks

Robust and imperceptible image watermarking by DC coefficients using singular value decomposition

Abstract: Main problem frequently encountered in all schemes transform domain watermarking technique is the robustness and imperceptibility. Due to achieved optimal result most algorithms of image watermarking using combination two or more transformation domain. This paper proposed Discrete Cosine Transform (DCT) and Singular Value Decomposition (SVD) to embed binary watermark to color Image. Before the message embedded in the color images, we converting RGB to YCbCr color space. Luminance component will be split into subblock and it has been transformed by DCT to produce DC and AC coefficients. DC coefficients selected as embedding place because it is perceptually usefulness and robust against various attacks. DC coefficients will be collected from every sub-block to create a reference image. Then apply SVD on reference image and embed message in singular values. Various attacks have been implemented and tested due to achieve robustness using Normalized Cross Correlation (NCC) and imperceptibility tested using Peak Signal to Noise Ratio (PSNR). High values of the measurement results show the feasibility of the proposed method. A highest PSNR value resulted 42.3009 dB, whereas a highest NCC values 0.9993 after JPEG Compression.

Hybrid method using HWT-DCT for image watermarking

Abstract: This paper proposed a hybrid method that had been solved both of these issues. We proposed a combination of Discrete Cosine Transform (DCT) and Haar Wavelet Transform (HWT) due to achieved imperceptibility and robustness comes from PSNR and NC value. Firstly, HWT applied in host image and then selected subband LL. LL subband will be divided into an 8×8 matrix in which each matrix was applied using DCT. Copyright embedded at the DC coefficient. In this paper we choose PSNR to measure the imperceptibility. Meanwhile, to measure the extracted copyright we used correlation coefficient. The results of this simulation method getting high PSNR value with an average of 43.2392 dB. Experiment results of extraction has tested using image watermarking attacks are JPEG compression, low pass filtering, mid filtering, unsharp filter, gaussian noise, cropingp, blurring, salt and pepper with the average value of the correlation coefficient of 0.8676. Our results proved that HWT-DCT has fulfilled the purpose of image watermarking, namely copyright protection.

Image watermarking using low wavelet subband based on 8×8 sub-block DCT

Abstract: Currently, digital image watermarking is widely implemented as a tool for copyright protection. Along with technological developments, the technique of theft and manipulation of digital imagery increasingly sophisticated. So research on image watermarking should continue to grow for improved copyright security. This paper proposes the use of Discrete Cosine Transform (DCT) based on 8×8 sub-blocks to convert the original image from spatial to the frequency domain. The DCT transform result will produce AC and DC coefficients. Next, the DC coefficients are collected into a matrix to be transformed with DWT. The result of DWT transformation yield four subbands. Then the LL subband is selected for embedding copyright in the form of a binary image with a certain watermark strength value. The final step inverse the DWT-DCT transform to produce a watermarked image. From the results of this study, high imperceptibility values are evidenced by PSNR and MSE. Image watermark toughness is also measured by NC and is proven to be robust from various attacks mainly from JPEG compression attacks.

An improved secure image hiding technique using PN-sequence based on DCT-OTP

Abstract: The security of sending confidential data on the internet is very important. Thus, the data security techniques must be developed. Techniques like cryptography and steganography are techniques that are widely used to protect information messages that are confidential. Steganography is a technique to hide, while cryptography is a technique to encode. The combination of the two techniques produces a better security system. This paper proposed a combination of steganography and cryptography by using Discrete Cosine Transform (DCT), One Time Pad (OTP) and PN-Sequence implemented on a digital image. The test and measurement method used to determine the quality of stego image or final image is using Peak Signal to Noise Ratio (PSNR) and Mean Square Error (MSE). On the other hand, Normalized Correlation Coefficient (NCC) is used to measure the quality of extraction and decryption messages from the proposed StegoCrypt method. The experimental results of this study get better results compared to previous studies where this technique has better imperceptibility and is more resistant to JPEG compression, median filter, and crop attacks.

StegoCrypt method using wavelet transform and one-time pad for secret image delivery

Abstract: In this study, the StegoCrypt technique is proposed using a combination of Discrete Wavelet Transform (DWT) and One-Time Pad (OTP). Cover image with size 512 ∗ 512 transformed with Wavelet transformation of four levels. For the first level to the third level, the subband LL is selected to obtain LL3 subband. At fourth level, the LL3 is then transformed into HH4 subband with help of wavelet transformation. This is done to gain strength and imperceptibility to the stego image. The secret message that is used is a binary image with a size of 32 ∗ 32. Secret image is encrypted with OTP before it is inserted in the host image. To test the quality of imperceptibility, stego images were measured using PSNR and MSE. While the quality of secret image reconstruction of extraction and decryption results are measured by using NCC. Robustness of stego image is also tested with JPEG compression attacks. From the test results can be concluded that the proposed method works well and get a better quality stego image. The secret image reconstruction results are also perfect as well as robust to JPEG compression attacks.

A secure image watermarking using Chinese remainder theorem based on haar wavelet transform

Abstract: This study proposed concealment of information through watermarking techniques using a combination of Haar Wavelet Transform (HWT) and Chinese Remainder Theorem (CRT) algorithms. The use of HWT can improve the imperceptibility of the watermarked image. While the CRT has a shorter operation and minimal distortion and provides security to copyright because of its insertion technique that indirectly also encrypts the copyright itself. Watermark insertion is done by using HWT on subband LL followed by CRT process. The evaluation standard that will be used in this research is Peak Signal to Noise Ratio (PSNR) and Normalized Correlation (NC). From the measurement results obtained the average value of PSNR 63.55dB, which shows the very high imperceptible quality of the watermarked image. While copyright extraction can be recovered perfectly with the value of NC 1.

Sphinx4 for Indonesian continuous speech recognition system

Abstract: Automatic Speech Recognition (ASR) is a technology which is capable to convert speech into text. Research in this field is growing very rapidly and is applied in multiple languages. Voice recognition of isolated word and connected word for Indonesian language has been done with various approaches. This is done to get better accuracy in recognition. However, there are still a few research about introduction of continuous speech for the Indonesian language. This paper describes the efforts made to build an Indonesian automatic speech recognition to recognize continuous speech using Sphinx4 (toolkit from CMUSphinx). There are three steps taken to build Indonesia ASR, those are preparing corpus, forming acoustic model and testing. The result of the acoustic model test which was formed showed the value of 23% word error rate and 32,8% sentence error rate. The lower the two variables, the better the introduction to the input given speech files.

Perlindungan hak cipta pada citra digital menggunakan least significant bit berbasis deteksi tepi canny

Abstract: Watermarking merupakan salah satu ilmu penyembunyian data dengan tujuan untuk mengamankan citra, salah satunya yaitu copyright protection . Beberapa algoritma dalam watermarking yang dikenal handal pada domain spasial salah satunya Least Significant Bit (LSB) yang diketahui telah diimplementasikan pada steganografi dan kriptografi. LSB mempunyai kelebihan dalam hal imperceptibility , dengan kata lain pada citra hasil embedding maupun extracting dengan citra asli tidak terdapat perbedaan secara kasat mata. Untuk meningkatkan imperceptibility ke dalam tingkat yang lebih tinggi maka LSB akan dikombinasikan dengan Canny . Kelebihan Canny yaitu tidak merubah persepsi dari citra yang diolah. Pada makalah ini, eksperimen dilakukan menggunakan citra grayscale berukuran 256x256 sejumlah masing-masing 4 citra berformat *.bmp dengan pesan yang akan disisipkan berupa file *.txt. Hasil dari eksperimen menunjukkan nilai PSNR yang tinggi, dimana nilai PSNR hasil embedding dengan Canny hanya 53,8370 dB, sedangkan dengan mengimplementasikan LSB-Canny didapat nilai PSNR tertinggi 75,6510 dB. Perbedaan yang cukup jauh tersebut diakibatkan oleh penentuan lokasi embedding yang dilakukan oleh Canny . Perolehan nilai MSE yang mendekati 0 maka semakin bagus kualitas citra yang dihasilkan. Nilai MSE terbaik yang dihasilkan oleh LSB yaitu 0,2688 sedangkan pada LSB-Canny nilai MSE terbaik yaitu 0,0018.