CASIA V1 Research Articles

Iris Recognition based on Statistically Bound Spatial Domain Zero Crossing and Neural Networks

Purpose The iris pattern is an important biological feature of the human body. The recognition of an individual based on an iris pattern is gaining more popularity due to the uniqueness of the pattern among the people. Iris recognition systems have received attention very much due to their rich iris texture which gives robust standards for identifying individuals. Notwithstanding this, there are several challenges in unrestricted recognition environments. Methods This article discusses a highly error-resistant technique to implement a biometric recognition system based on the iris portion of the human eye. All iris recognition algorithms of the current day face a major problem of localization errors and the enormous time involved in this localization process. Spatial domain zero crossing may be the simplest and least complex method for localization. Yet, it has not been used due to its high sensitivity to erroneous edges, as a consequence of which more complex and time-consuming algorithms have taken its place. Appropriate statistical bounds imposed on this process help this method to be the least erroneous and time-consuming. Errors were reduced to 0.022% using this approach on the CASIA v1 & v2 datasets. Time consumption in this stage was the least compared to other algorithms. At the comparison stage, most algorithms use multiple comparisons to account for translation and rotation errors. This is time-consuming and very resource-hungry. Results The current approach discusses a robust method based on a single comparison, which works with a correct recognition of over 99.78% which is clearly demonstrated by tests. Conclusions The technique is to use a neural network trained to recognize special statistical and regional parameters unique to every person’s iris. The algorithm also gives sufficient attention to consider illumination errors, elliptical pupils, excess eyelash errors and bad contrast.

Convolutional neural network-based techniques and error level analysis for image tamper detection

<span>Photographs are the foremost powerful and trustworthy media of expression. At present, digital pictures not only serve forged information but also disseminate deceptive information. Users and experts with various objectives edit digital photographs. Images are frequently used as proof of reality or fact, therefore fake news or any publication that makes use of photos that have been altered in any way has a larger chance of deceiving readers. There is a need for a high-resolution image analysis model that processes individual pixels in images and a substantial amount of diverse image data, to detect image falsification. Convolutional neural network (CNN) with error level analysis (ELA) adopted in this research is found to be an ideal deep learning concept for detecting image manipulation. The model exhibited a validation accuracy of 99.6%, 99.7%, and 99.4% for CASIA V1.0, CASIA V2.0 and MICC datasets respectively. The accuracy for handmade tampered images was found to be 99.2%.</span>

Image Forgery Detection Based on Parallel Convolutional Neural Networks

Abstract— Due to the availability of deep networks, progress has been made in the field of image recognition. Images and videos are spreading very conveniently and with the availability of strong editing tools the tampering of digital content become easy. To detect such scams, we proposed techniques. In our paper, we proposed two important aspects of employing deep convolutional neural networks to image forgery detection. We first explore and examine different preprocessing method along with convolutional neural networks (CNN) architecture. Later we evaluated the different transfer learning for pre-trained ImageNet(via-fine-tuning) and implement it over our dataset CASIA V2.0. So, it covers the pre-processing techniques with basic CNN model and later see the powerful effect of the transfer learning models. Keywords— image tampering, convolution neural network (CNN), error level analysis (ELA), transfer learning, sharpening filter, fine-tuning

A Deep Learning Model for Splicing Image Detection

With the advancement of digital technology, manipulating images has become relatively easy through many photo editing techniques. One of the techniques is the splicing image method, which crops parts of images and puts them into another image creating a new composite image. The image splicing detection system is soon regarded as an exciting topic for many researchers to solve the problems of forgery images on the Internet, especially in social networks. ResNet-50 and VGG-16 are powerful architectures of convolutional neural networks, but they reveal many weaknesses when operating on low-end computers. The ultimate goal of this research is to create a model for image splicing detection working well in limited memory machines. The study proposes the model, which is the improvement of VGG-16 applying residual network (ResNet). As a result, the proposed model achieves a test accuracy of 92.5% while the ResNet-50 gives an accuracy of 85.6% after 20 epochs of training 9,319 images from the CASIA v2.0 dataset, which are used for forgery classification. The result proves the efficiency of the proposed model for image splicing detection, especially when working on low-end computers.

Image Splicing Detection Based on Discrete Wavelet Transform and co-occurrence Matrix

In this paper a method to determine whether an image is forged (spliced) or not is presented. The proposed method is based on a classification model to determine the authenticity of a tested image. Image splicing causes many sharp edges (high frequencies) and discontinuities to appear in the spliced image. Capturing these high frequencies in the wavelet domain rather than in the spatial domain is investigated in this paper. Correlation between high-frequency sub-bands coefficients of Discrete Wavelet Transform (DWT) is also described using co-occurrence matrix. This matrix was an input feature vector to a classifier. The best accuracy of 92.79% and 94.56% on Casia v1.0 and Casia v2.0 datasets respectively was achieved. This performance is reached when Cb and Cr channels in YCbCr color space only are utilized to form this feature vector. The outcomes of the experiments showed that the proposed method achieved very good results in comparison with existing splicing detection approaches.

Study the effect of eye diseases on the performance of iris segmentation and recognition using transfer deep learning methods

A new deep learning-based iris recognition system is presented in the current study in the case of eye disease. Current state of art iris segmentation is either based on traditional low accuracy algorithms or heavy-weight deep-based models. In the current study segmentation section, a new iris segmentation method based on illumination correction and a modified circular Hough transform is proposed. The current method also performs a post-processing step to minimize the false positives. Besides, a ground truth of iris images is constructed to evaluate the segmentation accuracy. Many deep learning models (GoogleNet, Inception_ResNet, XceptionNet, EfficientNet, and ResNet50) are applied through the recognition step using the transfer learning approach. In the experiment part, two eye disease-based datasets are used. 684 iris images of individuals with multiple ocular diseases from the Warsaw BioBase V1 and 1,793 iris images from the Warsaw BioBase V2 are also used. The CASIA V3 Interval Iris dataset, which contains 2,639 photographs of healthy iris, is used to train deep models once, and then the transfer learning of this normal-based eye dataset is used to retrain the same deep models using Warsaw BioBase datasets. Different scenarios for training and evaluating participants are used during experiments. The trained models are evaluated using validation accuracy, training time, TPR, FNR, PPR, FDR, and test accuracy. The best accuracies are 98.5% and 97.26%, which are recorded by the ResNet50 (2-layer of transfer learning) model trained on Warsaw BioBase V1 and V2, respectively. Results indicate that the effect of eye diseases is concentrated on the segmentation phase. For recognition, no significant impact is recognized. Some disease that affects the structure (bloody eyes, trauma, iris pigment) can affect the iris recognition step partially. Our study is compared with similar studies in the case of eye diseases. The comparison proves the efficiency and high performance of the proposed methodology against all previous models on the same iris datasets.

Dual-branch multi-scale densely connected network for image splicing detection and localization

In the last two decades, numerous methods have been proposed for image tampering forensics, especially when deep learning is leveraged as a powerful feature extraction technique. However, existing deep learning-based methods are not efficient to jointly explore robust representations of tampering traces from both low-level forensics patterns and high-level forensics semantics. Besides, they become unreliable for spliced regions with unknown scales in the evaluation stage. To overcome the above-mentioned limitations, a novel Dual-branch Multi-scale Densely Connected Network (DMDC-Net) is proposed in this work for image splicing detection. DMDC-Net is constructed based on an encoder-decoder framework, which equips a multi-scale dense (MD) module for feature reuse and pixel-wise feature fusion (PFF) module to leverage forensics clues from different data components adaptively. In DMDC-Net, the given input image will be fed into two individual subnets to obtain deep representation maps from RGB and high-frequency components, which contains both visual and noise inconsistencies for further analysis. Besides, the MD module is applied in Encoder of each subnet to make the model jointly learn low-level forensics patterns and high-level semantic representations, which incorporates local and global features by a multi-scale manner of dense connections. The outputs of each subnet are fused in the PFF module to take advantage of their complementarity by trainable weights and get the final detection score map. Experimental results demonstrate the effectiveness of our method and the superior performance than other state-of-the-art methods in CASIA v2.0 and Columbia dataset for different scenarios, such as post-processing operations.

Image Splicing Forgery Detection using Standard Division-Local Binary Pattern Features

Numerous aspects of daily life contribute to societal stability, and the security of people's perceptions of the world online is one target of various malicious attacks. Professional forgers can now quickly create copy-move, splice, or retouch photos with the use of today's advanced tools. It has been determined that splicing, is a widespread method of manipulating images. Image forgery can also lead to substantial setbacks and challenges, some of which may have significant ethical, moral, or legal consequences. Thus, the paper proposes a system that combines SD-LBP (Standard Devision-Local Binary Pattern) based passive picture splicing detection system and ANN classifier. The SD-LBP is created to have the benefits and avoid drawbacks of Local Binary Pattern (LBP). The SD-LBP extraction is typically performed by employing proposed SD value-based thresholding instead of the center pixel, which is robust to noise and other photometric attacks. The second part of the proposed system is the ANN classifier is used that extract the feature of images to lower the error and build a model that can tell spliced images from real photos that have been digitally altered. The proposed system is creating a reliable image forgery detection technique that was implemented with CASIA V2.0 standard dataset. The results showing that it outperformed compared with other methods on the in terms of accuracy (97.8%), sensitivity (98.6%), and specificity (97.1%). Most importantly, the proposed SFD method exceeded the state-of-the-art efforts in this field in terms of accuracy.

Lightweight MobileNet Model for Image Tempering Detection

In recent years, there has been a wide range of image manipulation identification challenges and an overview of image tampering detection and the relevance of applying deep learning models such as CNN and MobileNet for this purpose. The discussion then delves into the construction and setup of these models, which includes a block diagram as well as mathematical calculations for each layer. A literature study on Image tampering detection is also included in the discussion, comparing and contrasting various articles and their methodologies. The study then moves on to training and assessment datasets, such as the CASIA v2 dataset, and performance indicators like as accuracy and loss. Lastly, the performance characteristics of the MobileNet and CNN designs are compared. This work focuses on Image tampering detection using convolutional neural networks (CNNs) and the MobileNet architecture. We reviewed the MobileNet architecture's setup and block diagram, as well as its application to Image tampering detection. We also looked at significant literature on Image manipulation detection, such as major studies and their methodologies. Using the CASIA v2 dataset, we evaluated the performance of MobileNet and CNN architectures in terms of accuracy and loss. This paper offered an overview of the usage of deep learning and CNN architectures for image tampering detection and proved their accuracy in detecting manipulated images.

Deep Learning-Based Iris Segmentation Algorithm for Effective Iris Recognition System

In this study, a 19-layer convolutional neural network model is developed for accurate iris segmentation and is trained and validated using five publicly available iris image datasets. An integrodifferential operator is used to create labeled images for CASIA v1.0, CASIA v2.0, and PolyU Iris image datasets. The performance of the proposed model is evaluated based on accuracy, sensitivity, selectivity, precision, and F-score. The accuracy obtained for CASIA v1.0, CASIA v2.0, CASIA Iris Interval, IITD, and PolyU Iris are 0.82, 0.97, 0.9923, 0.9942, and 0.98, respectively. The result shows that the proposed model can accurately predict iris and non-iris regions and thus can be an effective tool for iris segmentation.

Efficient, accurate and fast pupil segmentation for pupillary boundary in iris recognition

Iris recognition is a robust biometric system—user-friendly, accurate, fast, and reliable. This biometric system captures information in a contactless manner, making it suitable for use during the COVID-19 pandemic. Despite its advantages such as high security and high accuracy, iris recognition still suffers from pupil deformation, motion blur, eyelids blocking, reflection occlusion and eyelashes obscure. If the pupillary boundary is not accurately segmented, iris recognition may suffer tremendously. Moreover, reflections in iris image may lead to an incorrect pupillary boundary segmentation. The segmentation accuracy can also be affected and reduced because of the presence of an unwanted noise created by the motion blur effect in iris image. Additionally, the pupillary boundary might change from circular shape to uneven or irregular shape because of the interference and obstruction in pupil region. Therefore, this work is carried out to determine an accurate, efficient and fast algorithm for the segmentation of pupillary boundary. First, the iris image is pre-processed with Wiener filter. Next, the respective iris image is assigned with a specific threshold. After that, the pixel property in iris image is computed to determine the pupillary boundary coordinates which are acquired from the measured pixel list and area in iris image. Finally, morphological closing is used to remove reflections in the inner region of pupil boundary. All experiments are implemented with CASIA v4 database and Matlab R2020a.

Robust and Swift Iris Recognition at distance based on novel pupil segmentation

One of the most common and successful biometric frameworks is iris recognition, which has yielded promising results in systems of access control and identity authentication. Recent systems of iris recognition have concentrated on photos taken in non-ideal circumstances. Non-ideal imaging environments involve the capture of iris pictures in motion, at a distance, or under near-infrared (NIR) or visible wavelength illumination, which introduces noise factors such as gaze deviation, lack of focus, and obstruction by hair, eyelids, lighting, specular reflections, and eyeglasses. Segmenting irises is still a challenging task for eye images taken under non-ideal circumstances for iris recognition, and is strongly dependent on pupil segmentation; to estimate a line of sight using the pupil centre method, pupil segmentation is essential. The quality of eye pictures often changes due to noise effects and the individual variance between human eyes, making segmentation of the pupil a tricky problem. To address this issue, we propose an efficacious, swift, and robust iris recognition method that is convenient for eye pictures under non-ideal conditions. In this paper, we present a novel pupil segmentation method based on a fusion of the multi-scale gray-level co-occurrence matrix (MSGLCM) and the multi-range circle Hough transform (MRRCHT) methods. Using this segmentation approach, the pupil texture region is extracted correctly and a Hough transform is then used for the outer iris region. To support the improvements in segmentation accuracy, we also apply pre- and post-processing operations. An evaluation of the proposed method is conducted in this paper using the CASIA v4.0 (Distance) and MMU V2 databases, which reflect the challenges of iris recognition from non-ideal pictures. Extensive experimental findings clarify the effectiveness of the suggested method for non-ideal iris pictures.

Using XAI for Deep Learning-Based Image Manipulation Detection with Shapley Additive Explanation

In the arena of image forensics, detecting manipulations in an image is extremely significant because of the use of images in different fields. Various detection techniques have been suggested in the literature that are based on digging out the features from images to unveil the traces left by manipulation operations. In this paper, a deep learning-based approach is proposed in which a residual network is used to learn deep, complex features from preprocessed images for classification into authentic and forged images. There is statistical symmetry in similar types of images and asymmetry in different types of images. The proposed scheme can highlight the statistical asymmetry between authentic and forged images. In the proposed scheme, firstly, an RGB image is analyzed for different JPEG compression levels. The obtained difference between the error levels is used to extract enhanced LBP code. Then, the scale- and direction-invariant LBP (SD-LBP) code is transformed into SD-LBP feature maps to feed to a deep residual network. Next, the concept of explainable artificial intelligence (XAI) is used to help provide explanations and interpret the output, thereby raising the credibility of the proposed approach. The unique feature selection approach employed is the kernel SHAP method, which is focused on the Shapley values. This technique is used to pinpoint the specific characteristics that are responsible for the aberrant behavior of the forged images dataset. Later, the deep learning-based model is trained and validated using these feature sets. A pre-activation version of ResNet-50 architecture is used that achieved an accuracy of 99.31%, 99.52%, 98.05%, and 99.10% on CASIA v1, CASIA v2, IMD 2020, and DVMM datasets, respectively. The capability of the pretrained residual network and rich textural features, which are scale- and direction-invariant, helps to expand the detection accuracy of the proposed approach. The results confirmed that the method either produced competitive results or outperformed existing methods.

A Novel Deep Learning Architecture with Multi-Scale Guided Learning for Image Splicing Localization

The goal of image splicing localization is to detect the tampered area in an input image. Deep learning models have shown good performance in such a task, but are generally unable to detect the boundaries of the tampered area well. In this paper, we propose a novel deep learning model for image splicing localization that not only considers local image features, but also extracts global information of images by using a multi-scale guided learning strategy. In addition, the model integrates spatial and channel self-attention mechanisms to focus on extracting important features instead of restraining unimportant or noisy features. The proposed model is trained on the CASIA v2.0 dataset, and its performance is tested on the CASIA v1.0, Columbia Uncompressed, and DSO-1 datasets. Experimental results show that, with the help of the multi-scale guided learning strategy and self-attention mechanisms, the proposed model can locate the tampered area more effectively than the state-of-the-art models.

Recognition of human Iris for biometric identification using Daugman’s method

Iris identification is a well-known technology used to detect striking biometric identification procedures for recognizing human beings based on physical behaviour. The texture of the iris is unique and its anatomy varies from individual to individual. As we know, the physical features of human beings are unique, and they never change; this has led to a significant development in the field of iris recognition. Iris recognition tends to be a reliable domain of technology as it inherits the random variation of the data. In the proposed study of approach, we have designed and implemented a framework using various subsystems, where each phase relates to the other iris recognition system, and these stages are discussed as segmentation, normalisation, and feature encoding. The study is implemented using MATLAB where the results are outcast using the rapid application development (RAD) approach. We have applied the RAD domain, as it has an excellent computing power to generate expeditious results using complex coding, image processing toolbox, and high-level programing methodology. Further, the performance of the technology is tested on two informational groups of eye images MMU Iris database, CASIA V1, CASIA V2, MICHE I, MICHE II iris database, and images captured by iPhone camera and Android phone. The emphasis on the current study of approach is to apply the proposed algorithm to achieve high performance with less ideal conditions.

An improved low-complexity DenseUnet for high-accuracy iris segmentation network

Iris segmentation is one of the most important steps in iris recognition. The current iris segmentation network is based on convolutional neural network (CNN). Among these methods, there are still problems with the segmentation networks such as high complexity, insufficient accuracy, etc. To solve these problems, an improved low complexity DenseUnet is proposed to this paper based on U-net for acquiring a high-accuracy iris segmentation network. In this network, the improvements are as follows: (1) Design a dense block module that contains five convolutional layers and all convolutions are dilated convolutions aimed at enhancing feature extraction; (2) Except for the last convolutional layer, all convolutional layers output feature maps are set to the number 64, and this operation is to reduce the amounts of parameters without affecting the segmentation accuracy; (3) The solution proposed to this paper has low complexity and provides the possibility for the deployment of portable mobile devices. DenseUnet is used on the dataset of IITD, CASIA V4.0 and UBIRIS V2.0 during the experimental stage. The results of the experiments have shown that the iris segmentation network proposed in this paper has a better performance than existing algorithms.

Iris Features Extraction and Recognition based on the Scale Invariant Feature Transform (SIFT)

Iris Biometric authentication is considered to be one of the most dependable biometric characteristics for identifying persons. In actuality, iris patterns have invariant, stable, and distinguishing properties for personal identification. Due to its excellent dependability in personal identification, iris recognition has received more attention. Current iris recognition methods give good results especially when NIR and specific capture conditions are used in collaboration with the user. On the other hand, values related to images captured using VW are affected by noise such as blurry images, eye skin, occlusion, and reflection, which negatively affects the overall performance of the recognition systems. In both NIR and visible spectrum iris images, this article presents an effective iris feature extraction strategy based on the scale-invariant feature transform algorithm (SIFT). The proposed method was tested on different databases such as CASIA v1 and ITTD v1, as NIR images, as well as UBIRIS v1 as visible-light color images. The proposed system gave good accuracy rates compared to existing systems, as it gave an accuracy rate of (96.2%) when using CASIA v1 and (96.4%) in ITTD v1, while the system accuracy dropped to (84.0 %) when using UBIRIS v1.

PR‐NET: Progressively‐refined neural network for image manipulation localization

Current deep learning-based image manipulation localization methods achieve impressive performance when rich spatial features and information are fully utilized. However, most of them suffer from the irrelevance of semantic awareness when identifying various manipulation categories. This leads to false alarms on recognizing forged regions. In this paper, we propose a Progressively-Refined Neural Network (PR-Net), to localize the tampered regions progressively under a coarse-to-fine workflow. Specifically, PR-Net is composed of a Feature Extractor (FE) that captures feature intrinsic correlations and a Mask Generation Module (MGM) with three refining generators. The FE takes a CNN to extract the image features and introduces an attention mechanism Convolution Block Attention Module (CBAM) to suppress the image content and guide the extractor in exploring the inconsistencies between the manipulated and authentic regions. The MGM comprises three generators where the Coarse Mask RR-Generator generates a localization result roughly, the Candidate Mask RR-Generator generates a possible tampered region according to the rough localization measure, and the Fine Mask RR-Generator produces the final prediction of manipulated regions. We also utilize the Rotated Residual (RR) structure to suppress the image content during the generative process. The extensive experimental results on four benchmark data sets (NIST16, COVER, CASIA v1.0, and In-The-Wild) demonstrate the superior performance of PR-Net compared with the state-of-the-art methods in localizing the manipulated regions.

Performance Evaluation of New Feature based on Ordinal Pattern Analysis for Iris Biometric Recognition

The Iris recognition technique is currently the most efficient biometric identification system and is a common system on the practical front. Though most of the commercial systems use the patented Daugman’s algorithm, which mainly uses wavelet-based features, research is still active in identifying novel features that can provide personal identification. Here the first novel proposal of using ordinal pattern measure based on nonlinear time series analysis is put forth to characterize the unique pattern of the iris of individuals and thereby perform personal identification. Dispersion Entropy is a nonlinear time-series analysis method highly efficient in the characterization of the complexity of any data series with proven effectiveness in the characterization of model system dynamics as well as real-world data series. The results show that dispersion entropy can be used to identify iris images of specific individuals. The efficiency of this method is evaluated by computing correlation and RMSE between dispersion entropy values of normalized iris image rubber sheet data. The experimental results on the popular IRIS database- CASIA v1- demonstrate that the proposed method can effectively perform differential identification of iris images from different individuals. The results specifically indicate that the density of information along the angular direction of iris images which falls along the rows of rubber sheet data. This can be efficiently utilized with the method or ordinal pattern characterization and proves to be having promising potential for being incorporated into biometrics personal identification systems.

A Novel Image Splicing Detection Algorithm Based on Generalized and Traditional Benford’s Law

Due to the ease of access to platforms that can be used by forgers to tamper digital documents, providing automatic tools for identifying forged images is now a hot research field in image processing. This paper presents a novel forgery detection algorithm based on variants of Benford's law. In the proposed method, Mean Absolute Deviation (MAD) feature is extracted using traditional Benford's law. Also, generalized Benford's law is used for mantissa distribution feature vector. In addition to Benford's law-based features, other statistical features are used to construct the final feature vector. Finally, support vector machine (SVM) with three different kernel functions is used to classify original and forged images. The method has been tested on two common image datasets (CASIA V1.0 and V2.0). The experimental results show that 0.27% and 0.21% improvements on CASIA V1.0 and CASIA V2.0 datasets were achieved, respectively in detection accuracy by the proposed method in comparison to best state-of-the-art methods. The proposed efficient algorithm has a simple implementation. Moreover, on the basis of Benford’s law rich features are extracted from images so that classification process is efficiently performed by a simple SVM classifier in a short time.