Efficient Approach towards Detection and Identification of Copy Move and Image Splicing Forgeries Using Mask R-CNN with MobileNet V1

With the technological advancements of the modern era, the easy availability of image editing tools has dramatically minimized the costs, expense, and expertise needed to exploit and perpetuate persuasive visual tampering. With the aid of reputable online platforms such as Facebook, Twitter, and Instagram, manipulated images are distributed worldwide. Users of online platforms may be unaware of the existence and spread of forged images. Such images have a significant impact on society and have the potential to mislead decision-making processes in areas like health care, sports, crime investigation, and so on. In addition, altered images can be used to propagate misleading information which interferes with democratic processes (e.g., elections and government legislation) and crisis situations (e.g., pandemics and natural disasters). Therefore, there is a pressing need for effective methods for the detection and identification of forgeries. Various techniques are currently employed for the identification and detection of these forgeries. Traditional techniques depend on handcrafted or shallow-learning features. In traditional techniques, selecting features from images can be a challenging task, as the researcher has to decide which features are important and which are not. Also, if the number of features to be extracted is quite large, feature extraction using these techniques can become time-consuming and tedious. Deep learning networks have recently shown remarkable performance in extracting complicated statistical characteristics from large input size data, and these techniques efficiently learn underlying hierarchical representations. However, the deep learning networks for handling these forgeries are expensive in terms of the high number of parameters, storage, and computational cost. This research work presents Mask R-CNN with MobileNet, a lightweight model, to detect and identify copy move and image splicing forgeries. We have performed a comparative analysis of the proposed work with ResNet-101 on seven different standard datasets. Our lightweight model outperforms on COVERAGE and MICCF2000 datasets for copy move and on COLUMBIA dataset for image splicing. This research work also provides a forged percentage score for a region in an image.

Vision Sensor-Based Real-Time Fire Detection in Resource-Constrained IoT Environments

Fire detection and management is very important to prevent social, ecological, and economic damages. However, achieving real-time fire detection with higher accuracy in an IoT environment is a challenging task due to limited storage, transmission, and computation resources. To overcome these challenges, early fire detection and automatic response are very significant. Therefore, we develop a novel framework based on a lightweight convolutional neural network (CNN), requiring less training time, and it is applicable over resource-constrained devices. The internal architecture of the proposed model is inspired by the block-wise VGG16 architecture with a significantly reduced number of parameters, input size, inference time, and comparatively higher accuracy for early fire detection. In the proposed model, small-size uniform convolutional filters are employed that are specifically designed to capture fine details of input fire images with a sequentially increasing number of channels to aid effective feature extraction. The proposed model is evaluated on two datasets such as a benchmark Foggia’s dataset and our newly created small-scaled fire detection dataset with extremely challenging real-world images containing a high-level of diversity. Experimental results conducted on both datasets reveal the better performance of the proposed model compared to state-of-the-art in terms of accuracy, false-positive rate, model size, and running time, which indicates its robustness and feasible installation in real-world scenarios.

Dual Attention Multiscale Network for Vessel Segmentation in Fundus Photography

Background: Automatic vessel structure segmentation is an essential step towards an automatic disease diagnosis system. The task is challenging due to the variance shapes and sizes of vessels across populations.Methods: A multiscale network with dual attention is proposed to segment vessels in different sizes. The network injects spatial attention module and channel attention module on feature map which size is 1 8 of the input size. The network also uses multiscale input to receive multi-level information, and the network uses the multiscale output to gain more supervision. Results: The proposed method is tested on two publicly available datasets: DRIVE and CHASEDB1. The accuracy, AUC, sensitivity, specificity on DRIVE dataset is 0.9615, 0.9866, 0.7693, and 0.9851, respectively. On the CHASEDB1 dataset, the metrics are 0.9797, 0.9895, 0.8432, and 0.9863 respectively. The ablative study further shows effectiveness for each part of the network. Conclusions: Multiscale and dual attention mechanism both improves the performance. The proposed architecture is simple and effective. The inference time is 12ms on a GPU and has potential for real-world applications. The code will be made publicly available.

Scaling the U-net: segmentation of biodegradable bone implants in high-resolution synchrotron radiation microtomograms

Highly accurate segmentation of large 3D volumes is a demanding task. Challenging applications like the segmentation of synchrotron radiation microtomograms (SRμCT) at high-resolution, which suffer from low contrast, high spatial variability and measurement artifacts, readily exceed the capacities of conventional segmentation methods, including the manual segmentation by human experts. The quantitative characterization of the osseointegration and spatio-temporal biodegradation process of bone implants requires reliable, and very precise segmentation. We investigated the scaling of 2D U-net for high resolution grayscale volumes by three crucial model hyper-parameters (i.e., the model width, depth, and input size). To leverage the 3D information of high-resolution SRμCT, common three axes prediction fusing is extended, investigating the effect of adding more than three axes prediction. In a systematic evaluation we compare the performance of scaling the U-net by intersection over union (IoU) and quantitative measurements of osseointegration and degradation parameters. Overall, we observe that a compound scaling of the U-net and multi-axes prediction fusing with soft voting yields the highest IoU for the class “degradation layer”. Finally, the quantitative analysis showed that the parameters calculated with model segmentation deviated less from the high quality results than those obtained by a semi-automatic segmentation method.

Sampling-based approximate skyline calculation on big data

Nowadays, big data is coming to the force in a lot of applications. Processing a skyline query on big data in more than linear time is by far too expensive and often even linear time may be too slow. It is obviously not possible to compute an exact solution to a skyline query in sublinear time, since an exact solution may itself have linear size. Fortunately, in many situations, a fast approximate solution is more useful than a slower exact solution. This paper proposes two sampling-based approximate algorithms for processing skyline queries. The first algorithm obtains a fixed size sample and computes the approximate skyline on it. The error of the algorithm is not only relatively small in most cases, but also is almost unaffected by the input size. The second algorithm returns an [Formula: see text]-approximation for the exact skyline efficiently. The running time of the algorithm has nothing to do with the input size in practical, achieving the goal of sublinearity on big data. Experiments verify the error analysis of the first algorithm, and show that the second is much faster than the existing skyline algorithms.

Coresets for the Average Case Error for Finite Query Sets

Coreset is usually a small weighted subset of an input set of items, that provably approximates their loss function for a given set of queries (models, classifiers, hypothesis). That is, the maximum (worst-case) error over all queries is bounded. To obtain smaller coresets, we suggest a natural relaxation: coresets whose average error over the given set of queries is bounded. We provide both deterministic and randomized (generic) algorithms for computing such a coreset for any finite set of queries. Unlike most corresponding coresets for the worst-case error, the size of the coreset in this work is independent of both the input size and its Vapnik–Chervonenkis (VC) dimension. The main technique is to reduce the average-case coreset into the vector summarization problem, where the goal is to compute a weighted subset of the n input vectors which approximates their sum. We then suggest the first algorithm for computing this weighted subset in time that is linear in the input size, for n≫1/ε, where ε is the approximation error, improving, e.g., both [ICML’17] and applications for principal component analysis (PCA) [NIPS’16]. Experimental results show significant and consistent improvement also in practice. Open source code is provided.

On Self-Aggregations of Min-Subgroups

Preservation of structures under aggregation functions is an active area of research with applications in many fields. Among such structures, min-subgroups play an important role, for instance, in mathematical morphology, where they can be used to model translation invariance. Aggregation of min-subgroups has only been studied for binary aggregation functions . However, results concerning preservation of the min-subgroup structure under binary aggregations do not generalize to aggregation functions with arbitrary input size since they are not associative. In this article, we prove that arbitrary self-aggregation functions preserve the min-subgroup structure. Moreover, we show that whenever the aggregation function is strictly increasing on its diagonal, a min-subgroup and its self-aggregation have the same level sets.

Realtime Object Detection Masa Siap Panen Tanaman Sayuran Berbasis Mobile Android Dengan Deep Learning

Determining the harvesting period can be done visually, physically, computationally, and chemically. Since the harvesting process is crucial, late harvesting will affect post-harvest and production quality. Leafy vegetables have a relatively short ready-to-harvest period. Visual recognition of the harvesting period combined with image processing can recognize harvesting vegetables' visual characteristics. This study aims to build a deep learning-based mobile model to detect real-time vegetable plant objects such as bok choy, spinach, kale, and curly kale to determine whether these vegetables are ready for harvest. Mobile-based architecture is chosen due to latency, privacy, connectivity, and power consumption reason since there is no round-trip communication to the server. In this research, we use MobileNetV3 as the base architecture. To find the best model, we experiment using different image input size. We have obtained a maximum MAP score of 0. 705510 using a 36,000 image dataset. Furthermore, after implementing the model into the Android mobile application, we analyze the best practice in using the application to capture distance. In real-time detection usage, the detection can be done with an ideal distance of 5 cm and 10 cm.

COVID-19 Diagnosis from Chest X-Ray Images Using a Robust Multi-Resolution Analysis Siamese Neural Network with Super Resolution Convolutional Neural Network: Algorithm Development and Validation (Preprint)

BACKGROUND The chest x-ray (CXR) is becoming a useful method in the evaluation of coronavirus disease-19 (COVID-19). Despite the global COVID-19 uprising, utilizing computer-aided diagnosis approach for COVID-19 classification based on CXR images could significantly reduce clinician burden. There is no dispute that low resolution, noisy and irrelevant annotations in chest x-ray images is a major constraint to the performance of AI-based COVID-19 diagnosis. While few studies have made huge progress, they underestimate these bottlenecks. OBJECTIVE In this study, we propose a Super Resolution based Siamese Wavelet Multi-Resolution Convolutional Neural Network called COVID-SRWCNN for COVID-19 Classification using chest x-ray images. METHODS Concretely, we first reconstruct high-resolution (HR) counterparts from low resolution (LR) images of CXR images in order to enhance the quality of the dataset for improved performance of our model by proposing a novel enhanced fast super resolution convolutional neural network (EFSRCNN) to capture texture details in each given chest x-ray image. Since the datasets are collected from different sources with varying resolutions and the input layer of a convolutional neural network requires that the input size of the images in the training distribution must be fixed, therefore we extend the super resolution convolutional neural network by introducing an adaptive scaling operation to resize the images to a fixed resolution prior to the enhancement operation. Exploiting a mutual learning approach, the HR images are passed to the proposed siamese wavelet multi-resolution convolutional neural network to learn the high-level features for COVID-19 classification. RESULTS We validate the proposed COVID-SRWCNN model on public-source datasets achieving an accuracy of 99.6%, precision of 99.7%, and F1 score of 99.9%. Our screening technique achieved 99.8 % AUC, 99.7% sensitivity and 99.6% specificity. CONCLUSIONS Owing to the fact that COVID-19 chest x-ray dataset are low in quality, experimental results show that our proposed algorithm obtained up-to-date performance which is useful for COVID-19 screening.