Digital Image Identification Research Articles

A Digital Image Steganographic Detection Method for LSB Steganography

With the advancement of communication technology, communication information can no longer be transmitted merely through text but can be covertly conveyed via various types of communication carriers like images, audio, and video. This brings forth new challenges to traditional detection techniques that mainly target text information. Digital image steganalysis aims to detect and uncover communication information by recognizing steganographic behaviors contained within digital image files. This paper takes digital image steganographic activities based on least significant bit (LSB) as the main research object and conducts analysis and identification of digital images with steganographic information through visual attack method. It has been verified through experiments that steganographic activities based on LSB are hardly detectable through direct visual observation, but can be identified visually after undergoing visual attack processing.

Digital Image Forensics Using Deep Learning and Person Identification

In the realm of computer vision and image processing, the task of digital image identification holds significant importance across various domains. This paper presents a comparative study of two distinct methodologies for digital image identification: deep learning and the Haar cascade algorithm. Deep learning, specifically Convolutional Neural Networks (CNNs), has emerged as a powerful tool for automatically learning hierarchical representations from data and has achieved remarkable success in image-related tasks. In contrast, the Haar cascade algorithm, a classic machine learning technique, offers real-time object detection capabilities with its efficient feature-based approach. Through a series of experiments and evaluations on benchmark datasets, we analyze the performance, strengths, and limitations of these methodologies. Factors such as dataset size, computational resources, and application requirements are considered in the comparison. Our findings provide insights into the suitability of deep learning and the Haar cascade algorithm for various image identification tasks, aiding practitioners and researchers in selecting the most appropriate approach based on specific project needs. This research contributes to advancing the field of image processing and computer vision by offering a comprehensive analysis of these two prominent methodologies in digital image identification. Key Words: Deep Learning, Convolutional Neural Network (CNN), Haar cascade algorithm, image, forgery, detection

Identification Of Hijaiyah Letters Image Using Extreme Learning Machine Method

The development of digital image processing technology has many benefits, one of which is the identification of an object, such as the identification of the image of the hijaiyah letter. Hijaiyah letters are the letters of the arabic alphabet as the original language of the Qur'an. In essence, humans have the ability to recognize and distinguish hijaiyah letter patterns from one another, but this is not the case with computers, using digital images and machine learning, in this study an identification concept was built by recognizing the image of hijaiyah letters using one of the machine learning methods. namely the extreme learning machine (ELM) method. Extreme learning machine (ELM) is a feedforward neural network with one hidden layer or better known as single hidden layer feedforward neural networks (SLFNs). Therefore, the purpose of this study is that the computer can identify objects as well as human capabilities and see how accurate the results obtained in the ELM method are. The digital image identification process using the extreme learning machine (ELM) method is carried out in two stages, namely training and testing, where previously the preprocessing process was carried out first by changing the color of the RGB image to HSV and processing the color v, then segmentation was carried out with the aim of separating the objects (foreground) with the background, then to make it easier to recognize the pattern, a morphological process is carried out. From the simulation carried out on the test data, the results obtained an average accuracy of 90% with an error of 10%.

SNRCN2: Steganalysis noise residuals based CNN for source social network identification of digital images

Identifying the processing history of digital images related to source acquisition device, image manipulations, and Source Social Media Network (SSMN) identification is important for a forensic analyst to verify image source, trustworthiness, and integrity. Nowadays, social media networks have become a major medium of image sharing and identifying the SSMN of digital images has allured attention in the image forensic scientific community. With the precedence of deep Convolutional Neural Networks (CNNs) in the domain of image forensics, we propose a novel Steganalysis Noise Residuals based CNN (SNRCN2) for digital images SSMN identification. Inspired by the fact that image content can heavily obscures the artifacts of post-processing operations, the proposed CNN utilizes the features from steganalysis-based noise residuals to highlight the artifacts introduced by social media networks. Therefore, the given input image is high-pass filtered using well-known 30 Spatial Rich Model (SRM) filters to obtain noise residuals. Afterward, these noise residuals are passed to an efficient CNN for the extraction of high-level features related to social media networks. Lastly, a fully-connected layer is used for classification purposes. The experimental results show that the proposed model outperforms the existing techniques by providing an image-level accuracy of 99.53% and 100% on the VISION and Forchheim datasets, respectively. To further evaluate the robustness of the proposed model, we create a combined dataset that includes the images of common classes from both datasets. The results of the combined dataset further confirm the efficacy of the proposed model.

Digital Image Identification and Verification Using Maximum and Preliminary Score Approach with Watermarking for Security and Validation Enhancement

Digital face approaches possess currently received awesome attention because of their huge wide variety of digital audio, and visual programs. Digitized snapshots are progressively more communicated using an un-relaxed medium together with cyberspace. Consequently, defence, clinical, medical, and exceptional supervised photographs are essentially blanketed towards trying to employ it; such controls ought to damage such choices constructed totally based on those pictures. So, to shield the originality of digital audio/visual snapshots, several approaches proposed. Such techniques incorporate traditional encoding, breakable and nominal breakable watermarking with virtual impressions which are based upon the material of image content. Over the last few decades, various holistic approaches are proposed for improving image identification and verification. In this paper, a combination of both the feature level and score level of different techniques were used. Image is one of the identities of a person which reflects its emotions, feeling, age etc. which also helps to gather an information about a person without knowing their name, caste, and age and this could be not of much importance when it is used for domestic or framing applications. To secure the originality of digital audio/visual impressions many methods come into pictures and are proposed which include digital signatures, watermarking, cryptography, and fragile depend upon face contents. The objective of this research article is to identify & verify real-time video images using feature and score levels using watermarking that will help to judge the authenticity of any images at the initial stage by extracting the features which are evaluated by following an algorithm known as Viterbi and where input data is changed initially into an embedded treat or state then the matrix is evaluated of achieved transformation and on this basis preliminary score estimation will be generated after many iterations for each image that will help in validation. Finally, the tested image will be verified using several approaches to protect and provide security to the original image being verified. This approach may be useful for different surveillance applications for real-time image identification and verification. Also, measurement of accuracy was done by reconfiguring the HMM to identify the constant segmentation and feature removal of the image was settled by initializing parameters and teaching the image feature using the algorithm “Viterbi”.

A Leaky Integrate-and-Fire Neuron Based on Hexagonal Boron Nitride (h-BN) Monocrystalline Memristor

As a competitive candidate for artificial neurons, memristors have become the focus of intense research owing to their intrinsic ion migration tunability, enabling an authentic implementation of biomimicry. However, they still suffer from variability issues due to 3-D uncontrollable filament dynamics in an amorphous medium and modeling of switching dynamics underlying filament growth and rupture is still under investigation. In this work, we present volatile memristors that exhibit desired characteristics for neuromorphic computing with low performance variations utilizing a hexagonal boron nitride (h-BN) monocrystalline as a switching medium. Theoretical investigations assisted by the Monte Carlo simulation combined with experimentally detected <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${I}$ </tex-math></inline-formula> – <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${V}$ </tex-math></inline-formula> characteristics described that the electric field dominates the set process, whereas the Gibbs–Thomson interfacial energy minimization and heat dissipation influence the relaxation process mostly. Additionally, h-BN memristors with high switching uniformity provide an ideal hardware platform for credible neuron emulation and software identification of digital images.

The Identification of Beef and Pork Using Neural Network Based on Texture Features

The actual problem that frequently happens related to meat sales at conventional markets is the manipulation of pork and beef. It can happen as both visual textures bear resemblances. Texture is a crucial part of an object. In image processing, textures can be used for classification, recognition or prediction of an image. This paper offers the Minimum Overlap Probability - Neural Network method for the identification of digital image features of pork and beef.. Minimum Overlap Probability was employed to select features of the strongest characteristics, whilst Neural Network is used for training and classification. Based on the test results, the strongest features are maximum probability, contrast, sum average, autocorrelation, and energy and entropy sum. Based on MOP-NN Model test result, the digital image identification of beef and pork has performance with an accuracy of 96% on 400 images of sample data.

Dimension Reduction of Digital Image Descriptors in Neural Identification of Damaged Malting Barley Grains.

The paper covers the problem of determination of defects and contamination in malting barley grains. The analysis of the problem indicated that although several attempts have been made, there are still no effective methods of identification of the quality of barley grains, such as the use of information technology, including intelligent sensors (currently, quality assessment of grain is performed manually). The aim of the study was the construction of a reduced set of the most important graphic descriptors from machine-collected digital images, important in the process of neural evaluation of the quality of BOJOS variety malting barley. Grains were sorted into three size fractions and seed images were collected. As a large number of graphic descriptors implied difficulties in the development and operation of neural classifiers, a PCA (Principal Component Analysis) statistical method of reducing empirical data contained in the analyzed set was applied. The grain quality expressed by an optimal set of transformed descriptors was modelled using artificial neural networks (ANN). The input layer consisted of eight neurons with a linear Postsynaptic Function (PSP) and a linear activation function. The one hidden layer was composed of sigmoid neurons having a linear PSP function and a logistic activation function. One sigmoid neuron was the output of the network. The results obtained show that neural identification of digital images with application of Principal Component Analysis (PCA) combined with neural classification is an effective tool supporting the process of rapid and reliable quality assessment of BOJOS malting barley grains.

Digitization in Detecting Fish Freshness Quality: Literature Review

Fish contain a lot of nutrients that are very good for the body, but fish are often traded in a postmortem state with poor quality. One of the reasons is the difficulty of detecting the quality of fish freshness. Many studies currently make applications and special tools to see the fish quality in the digitalization of tools to facilitate human work. The method used is by identifying digital images, acoustic techniques, and electronic noses. In digital image identification, the fish’s eye image is applied by taking pictures of the condition of the fish on a smartphone device and then processing it using the small square method or curve fitting. After that, the identification of the acoustic form utilizes sound waves emitted to the observed target. In addition, the electric nose method is carried out by using the nose's working principle, which receives a response to odors that arise from rotting fish.

Identification of Formaldehyde Bananas using Learning Vector Quantization

Bananas that ripen with chemical process or do not ripen naturally usually, this can be recognized by the presence of blackish patches on the surface of the skin. But visual recognition has its drawbacks, which is that it is difficult to recognize similarities between formalin bananas and natural bananas, resulting in a lack of accurate identification. In this study, a system was built that can determined formalin bananas and natural bananas through digital image identification using supervised classification. The image to be identification previously goes through the process of transforming RGB (Red Green Blue) color to Grayscale, and the process of extracting texture features using statically recognizable features through histograms, in the form of average, standard deviation, skewness, kurtosis, energy, entropy and smoothness. The extraction of texture features is classified with LVQ (Learning Vector Quantization) to determine formalin or natural bananas. The test was conducted with 122 banana imagery sample data, 100 imagery as training data consisting of 50 imagery for natural bananas and 50 imagery for bananas formalin, 22 imagery as test data. The test results showed LVQ method has the best percentage at Learning Rate 0.1, Decreased Learning Rate 0.75 and maximum epoch of 1000 with the smallest epoch of 7, obtained accuracy 90.90%, precision 84.61% and recall 100%.

Model for identification of the background emission of Universe microwaves

This article discusses the identification of digital images in the map of the background emission of Universe microwaves developed on the basis of data from the European Space Agency telescope Planck upon applying a simulation of parameter z from the Doppler effect formula and the theory of covariance functions. The value of the Doppler parameter z shows an approximate mutual speed of two digital images (objects) with respect to the speed of light in vacuum. For a calculation of values of parameter z, vectors transformed by digital images and the pixels intensity of the white solar spectrum were used. The values of parameter z enabled calculation of the mutual speed of the digital images and the speeds of their movement with respect to the solar system. For calculation of estimates of the covariance functions, the algebraic additions of vectors transformed by the digital images and single vectors were applied. While varying the scale of a digital image, the frequencies of single pixel colours do not change, that is, they remain constant and no impact of the scale variation expresses itself in the calculation procedure of covariance functions. For the identification of images, the spectrum of RGB format was used. The average values of parameter z in the Doppler formula were calculated according to the formula created on the basis of expressions of cross-covariance functions of the algebraic addition of relevant transformed vectors and single vectors. The mutual speed of the movement of images of a part of the Universe was established. The calculation data were obtained using MATLAB-based computer programmes.

K-means clustering algorithm in image recognition tasks

The article is an overview of the scientific direction of graphic image recognition based on the k-means clustering algorithm. In particular, we have analyzed the clustering of data in digital image identification problems; we have described the operation principle of this algorithm with an illustration, where the number of selected clusters is equal to two and we presented a block diagram of the algorithm. We have proposed a scheme for implementing a machine vision system using this algorithm in the process of detecting overheated sockets on rail. We have performed color-based segmentation simulations over infrared images to detect high temperature sections in MATLAB. More specifically: two conditional images are taken, showing the railroad socket, one of which is normal and the other is overheated. The simulation results show that color-based segmentation allows accurate segmentation of thermal imaging. The application of this algorithm in the process of creating software for thermographic cameras for the purpose of recognizing temperature is effective, because this algorithm enables image compression, speeds up data processing, and enables the thermal performance of railroad socket to be monitored at high speeds.

Robust Perceptual Image Hashing using SIFT and SVD

With the advancement in technology, digital data such as image, video, etc. can be easily manipulated. Image hashing is a method that can be used for authentication and identification of digital images. In this communication, a robust image hashing technique is proposed using scale invariant feature transform (SIFT), singular value decomposition (SVD) and Zernike moment. Zernike moment is used to restore the image against the rotation attack. Potential points are generated from the image by using SIFT. Block processing of equal size is performed on the input image. Key points within the same block are used to generate hashing values. The experimental outcome shows that the proposed technique can withstand different types of attacks. Receiver operating characteristic curve comparison specifies that our method outperforms other existing methods under consideration.

Camera Source Identification of Digital Images Based on Sample Selection

With the advent of the Information Age, the source identification of digital images, as a part of digital image forensics, has attracted increasing attention. Therefore, an effective technique to identify the source of digital images is urgently needed at this stage. In this paper, first, we study and implement some previous work on image source identification based on sensor pattern noise, such as the Lukas method, principal component analysis method and the random subspace method. Second, to extract a purer sensor pattern noise, we propose a sample selection method to improve the random subspace method. By analyzing the image texture feature, we select a patch with less complexity to extract more reliable sensor pattern noise, which improves the accuracy of identification. Finally, experiment results reveal that the proposed sample selection method can extract a purer sensor pattern noise, which further improves the accuracy of image source identification. At the same time, this approach is less complicated than the deep learning models and is close to the most advanced performance.

Analysis of identification of digital images from a map of cosmic microwaves

This paper discusses identification of digital images from the cosmic microwave background radiation map formed according to the data of the European Space Agency “Planck” telescope by applying covariance functions and wavelet theory. The estimates of covariance functions of two digital images or single images are calculated according to the random functions formed of the digital images in the form of pixel vectors. The estimates of pixel vectors are formed on expansion of the pixel arrays of the digital images by a single vector. When the scale of a digital image is varied, the frequencies of single-pixel color waves remain constant and the procedure for calculation of covariance functions is not affected. For identification of the images, the RGB format spectrum has been applied. The impact of RGB spectrum components and the color tensor on the estimates of covariance functions was analyzed. The identity of digital images is assessed according to the changes in the values of the correlation coefficients in a certain range of values by applying the developed computer program.

Designing License Plate Identification through Digital Images with OpenCV

Abstract The aim of this research is to design and analyze the License Plate Identification program mediated through Digital Images or Automatic Number Plate Recognition (ANPR), especially by using desktop peripheral. In doing so, license plates attached, especially, on cars will be the test subject of this research. The ANPR is already implemented within the barrier gate parking system. It is able to record the data of the vehicles that come in yet it doesn’t necessarily recognize or identify the license plates installed on the vehicles. The fundamental goal of the ANPR program itself is actually to utilize digital image identification system in order to identify every single vehicle that goes in and out through the barrier gate parking system. From the result of the experimentation, the ANPR is able to detect and translate the license plates into a form of text in less than one second with 100% accuracy for high quality image, 82.6% for medium quality and 44.5% for low quality image. Series of analysis that the ANPR program situates involves; analysis on the ratio of the license plate, experimentation on the distance of license plate detection process, comparison between binary calculation methods using Adaptive Threshold and Global Threshold, and overall system examination. The result retrieved from the conducted analysis can be considered as a part of the ANPR system

Analysis of surface roughness parameters digital image identification

The article analyses some surface roughness parameters of metal parts determining the ability of the surface of digital image identification, covariance functions and Wavelet’s wave theory. Expressions of covariance functions are formed using random functions, made by spreading digital image pixel arrays by columns in the form of individual vectors. The digital images used for research may vary in scale, because the frequencies of colour waves with individual pixels remain constant in the images, therefore, the image change does not influence the scale in computing covariance functions. The colour spectrum of RGB format was applied to identify the surface images of the parts. There was analysed the influence of individual RGB colour tensor components on the estimates of digital image covariance functions. The identity of digital images was evaluated by the change of correlation coefficient values in the range of RGB colours. The software was applied to compute the above process.

Source Camera Identification Based on In-and-Inter Channel Traces

Source camera detecting, which is about establishing whether or not the images of interest are taken by the same camera, is a challenging problem. Most CFA(color filter array) interpolation algorithm characterize the feature of camera which take the image. In this paper, we propose an improved algorithm. Instead of using inter-channel demosaicking/color interpolation traces, we first extract four in-and-inter-channel variance maps, respectively, and then extract the shape, similarity and difference features of maps for camera model identification. The experimental results show that the method had significantly improved the accuracy rate of source digital image identification.

Multiscale digital image identification of heterogeneous elastic properties of softwoods

Multi-scale modelling of materials is now widely spread as it allows for a comprehensive analysis of their behaviour. The idea of this paper is to propose a methodology that couples Digital Images Correlation with multi-scale numerical simulations in order to identify a material behaviour that is too complex to obtain at the observation scale but that can be described in a simpler way at a finer scale. For this purpose, a Mechanical Image Correlation approach is proposed that involves a multi-scale numerical model. The capabilities of the proposed method is illustrated based on an example dedicated to the identification of heterogeneous orthotropic elastic properties of softwood.

A Visual Cryptography and Statistics Based Method for Ownership Identification of Digital Images

In this paper, a novel copyright protection scheme for digital images based on Visual Cryptography and Statistics is proposed. In our scheme, the theories and properties of sampling distribution of means and visual cryptography are employed to achieve the requirements of robustness and security. Our method does not need to alter the original image and can identify the ownership without resorting to the original image. Besides, our method allows multiple watermarks to be registered for a single host image without causing any damage to other hidden watermarks. Moreover, it is also possible for our scheme to cast a larger watermark into a smaller host image. Finally, experimental results will show the robustness of our scheme against several common attacks. Keywords—Copyright protection, digital watermarking, sampling distribution, visual cryptography.