Diverse Set Of Images Research Articles

Degrade for upgrade: Learning degradation representations for real-world low-light image enhancement

Most existing low-light image enhancement approaches predominantly focus on directly modeling the mapping between low-light images and normal-light images, but models trained on the single fixed data set often fail to achieve satisfactory results in complex real-world scenes. To address the limitations posed by limited data, this paper proposes a joint framework to learn spatially adaptive degradation and enhancement representations of bright-to-dark images and dark-to-bright images. Firstly, we design an illumination degradation model based on cycle consistent loss to learn how to degrade normal-light images into low-light reference images. The degradation model can synthesize a large number of normal/degraded-light pairs, which is beneficial for the training of the second-stage enhancement model. Subsequently, leveraging the self-luminance information of images, we construct a self-attention-based light enhancement model. The enhancement model is trained on mixed data, which combines both real normal/low-light image pairs and synthetic normal/degraded-light image pairs. In this way, the enhancement model can learn more features from a diverse set of images, enhancing its robustness. Extensive experiments on low-light image enhancement, low-light object detection, and nighttime image segmentation have demonstrated that our proposed method outperforms existing supervised and unsupervised methods across a variety of real-world scenes.

An automatic shoreline extraction method from SAR imagery using DeepLab-v3+ and its versatility

ABSTRACT This study presented an automatic shoreline extraction method from synthetic aperture radar (SAR) imagery using a semantic image segmentation model, DeepLab-v3+. Shorelines extracted from optical satellite images were employed as ground truth. More than 130,000 pairs of labeled images were generated, which substantially enhances the dataset’s objectivity for model training. By using different combination of SAR images from beaches with varying characteristics, several models were constructed. These models were then applied to 15 beaches of Japan to validate their accuracy and versatility. The developed machine learning model demonstrates high accuracy in shoreline extraction, particularly when trained with a diverse set of images from various beaches. This versatility is crucial for the model’s applicability across different coastal environments, underscoring the importance of incorporating a wide range of beach properties into the training dataset.

Improved vigenere using affine functions surrounded by two genetic crossovers for image encryption

This paper presents an improved method for encrypting color images, surpassing the effectiveness of genetic crossover and substitution operations. The technique incorporates dynamic random functions to enhance the integrity of the resulting vector, increasing temporal complexity to thwart potential attacks. The improvement involves integrating genetic crossover and utilizing two extensive pseudorandom replacement tables derived from established chaotic maps in cryptography. Following the controlled vectorization of the original image, our approach initiates with a first genetic crossover inspired by deoxyribonucleic acid (DNA) behavior at the pixel level. This genetic crossover is succeeded by a confusion-diffusion lap, reinforcing the connection between encrypted pixels and their neighboring counterparts. The confusion-diffusion process employs dynamic pseudorandom affine functions at the pixel level. Then a second genetic crossover operator is applied. Simulations conducted on a diverse set of images with varying sizes and formats showcase the robustness of our method against statistical, brute-force, and differential attacks.

Improved Vigenere approach incorporating pseudorandom affine functions for encrypting color images

This article presents an improvement to the traditional Vigenere encryption method, specifically adapted for the encryption of color images. This enhancement relies on the use of two chaotic maps widely employed in the field of cryptography. After vectorizing the original image and calculating the initialization value, which alters the seeding pixel to trigger the encryption process, our approach integrates two new large substitution tables. These tables are linked to confusion and diffusion functions, incorporating multiple reversible pseudo-random affine functions at the pixel level. Finally, a global permutation is applied to the entire resulting vector to increase the temporal complexity of potential attacks on our system. Simulations conducted on a diverse set of images of various sizes and formats demonstrate the resilience of our approach against any unexpected attacks.

Hair & Scalp Disease Detection Using Machine Learning & Image Processing

This study predicts hair disorders and provides tailored therapeutic suggestions using a deep learning model based on the VGG (Visual Geometry Group) architecture. A common dermatological problem, hair diseases can have a big impact on a person's physical and mental well-being. Early and accurate diagnosis as well as tailored treatment recommendations are essential for effective management. Inspired by the VGG design, we built a convolutional neural network (CNN) model in this study to analyze images of hair and scalp conditions. The model was trained using a vast and diverse set of images depicting various hair and scalp conditions. Transfer learning was used to modify the pre-trained VGG model so that it could recognize particular characteristics associated with particular hair issues. The prediction model consistently and reliably recognized a variety of hair problems that included dandruff, fungal infections, and alopecia. Reduction of false-positive and false negative outcomes in diagnosing diseases is dependent on high sensitivity and specificity. The suggested AI-based system has the potential to transform the dermatology field by providing prompt and accurate diagnosis of hair diseases and customized treatment recommendations. This study contributes to the ongoing efforts to use artificial intelligence and deep learning to improve healthcare outcomes, especially in the dermatology and skincare domains. Key Words: Convolutional Neural Network (CNN), Deep Learning, Hair Disorders, Visual Geometry Group (VGG).

Test-time Training for Hyperspectral Image Super-resolution.

The progress on Hyperspectral image (HSI) super-resolution (SR) is still lagging behind the research of RGB image SR. HSIs usually have a high number of spectral bands, so accurately modeling spectral band interaction for HSI SR is hard. Also, training data for HSI SR is hard to obtain so the dataset is usually rather small. In this work, we propose a new test-time training method to tackle this problem. Specifically, a novel self-training framework is developed, where more accurate pseudo-labels and more accurate LR-HR relationships are generated so that the model can be further trained with them to improve performance. In order to better support our test-time training method, we also propose a new network architecture to learn HSI SR without modeling spectral band interaction and propose a new data augmentation method Spectral Mixup to increase the diversity of the training data at test time. We also collect a new HSI dataset with a diverse set of images of interesting objects ranging from food to vegetation, to materials, and to general scenes. Extensive experiments on multiple datasets show that our method can improve the performance of pre-trained models significantly after test-time training and outperform competing methods significantly for HSI SR.

Symmetric Enhancement of Visual Clarity through a Multi-Scale Dilated Residual Recurrent Network Approach for Image Deraining

Images captured during rainy days present the challenge of maintaining a symmetrical balance between foreground elements (like rain streaks) and the background scenery. The interplay between these rain-obscured images is reminiscent of the principle of symmetry, where one element, the rain streak, overshadows or disrupts the visual quality of the entire image. The challenge lies not just in eradicating the rain streaks but in ensuring the background is symmetrically restored to its original clarity. Recently, numerous deraining algorithms that employ deep learning techniques have been proposed, demonstrating promising results. Yet, achieving a perfect symmetrical balance by effectively removing rain streaks from a diverse set of images, while also symmetrically restoring the background details, is a monumental task. To address this issue, we introduce an image-deraining algorithm that leverages multi-scale dilated residual recurrent networks. The algorithm begins by utilizing convolutional activation layers to symmetrically process both the foreground and background features. Then, to ensure the symmetrical dissemination of the characteristics of rain streaks and the background, it employs long short-term memory networks in conjunction with gated recurrent units across various stages. The algorithm then incorporates dilated residual blocks (DRB), composed of dilated convolutions with three distinct dilation factors. This integration expands the receptive field, facilitating the extraction of deep, multi-scale features of both the rain streaks and background information. Furthermore, considering the complex and diverse nature of rain streaks, a channel attention (CA) mechanism is incorporated to capture richer image features and enhance the model’s performance. Ultimately, convolutional layers are employed to fuse the image features, resulting in a derained image. An evaluation encompassing seven benchmark datasets, assessed using five quality metrics against various conventional and modern algorithms, confirms the robustness and flexibility of our approach.

VIDM: Video Implicit Diffusion Models

Diffusion models have emerged as a powerful generative method for synthesizing high-quality and diverse set of images. In this paper, we propose a video generation method based on diffusion models, where the effects of motion are modeled in an implicit condition manner, i.e. one can sample plausible video motions according to the latent feature of frames. We improve the quality of the generated videos by proposing multiple strategies such as sampling space truncation, robustness penalty, and positional group normalization. Various experiments are conducted on datasets consisting of videos with different resolutions and different number of frames. Results show that the proposed method outperforms the state-of-the-art generative adversarial network-based methods by a significant margin in terms of FVD scores as well as perceptible visual quality.

Object detection in hospital facilities: A comprehensive dataset and performance evaluation

Detecting objects in hospital indoor environments is critical for scene understanding and can have various applications in healthcare. Deep learning algorithms have proven to be effective in object recognition from images or videos, but the availability of annotated datasets plays a crucial role in their successful application. However, there is a shortage of datasets for object detection in hospital settings, hindering the advancement of hospital indoor object detection algorithms. In this paper, we present the Hospital Indoor Object Detection (HIOD) dataset, consisting of 4,417 images covering 56 object categories. The HIOD dataset represents the frequently encountered objects in hospitals and comprises 51,869 annotated objects. The dataset is characterized by dense annotation, with an average of 11.7 objects and 6.8 object categories per image. An object detection benchmark was established using the HIOD dataset and eight state-of-the-art object detectors. The benchmark provides a comprehensive evaluation of the performance of the selected object detectors on a large and diverse set of images of objects commonly seen in hospital environments. The results of the benchmark can be used to compare and analyze the performance of different object detectors and identify their strengths and weaknesses for use in hospital environments. In the benchmark, one-stage detectors have shown superior performance compared to two-stage detectors of similar parameter sizes. In particular, YOLOv6-L was able to attain a mean Average Precision (mAP) of 51.7% while operating at a detection speed of 255 FPS. The benchmark and dataset can serve as a valuable resource for researchers and practitioners in the field of computer vision and robotics, helping to advance the development of more effective and efficient object detection algorithms for developing automated operations in hospitals such as robotic disinfection and patient assistance.

A novel feature for monitoring the enzymatic harvesting process of adherent cell cultures based on lens-free imaging

Adherent cell cultures are often dissociated from their culture vessel (and each other) through enzymatic harvesting, where the detachment response is monitored by an operator. However, this approach is lacking standardisation and reproducibility, and prolonged exposure or too high concentrations can affect the cell’s viability and differentiation potential. Quantitative monitoring systems are required to characterise the cell detachment response and objectively determine the optimal time-point to inhibit the enzymatic reaction. State-of-the-art methodologies rely on bulky imaging systems and/or features (e.g. circularity) that lack robustness. In this study, lens-free imaging (LFI) technology was used to develop a novel cell detachment feature. Seven different donors were cultured and subsequently harvested with a (diluted) enzymatic harvesting solution after 3, 5 and 7 days of culture. Cell detachment was captured with the LFI set-up over a period of 20 min (every 20 s) and by optimising the reconstruction of the LFI intensity images, a new feature could be identified. Bright regions in the intensity image were identified as detaching cells and using image analysis, a method was developed to automatically extract this feature, defined as the percentage of detached cell regions. Next, the method was quantitatively and qualitatively validated on a diverse set of images. Average absolute error values of 1.49%, 1.34% and 1.97% were obtained for medium to high density and overconfluent cultures, respectively. The detachment response was quantified for all conditions and the optimal time for enzyme inhibition was reached when approximately 92.5% of the cells were detached. On average, inhibition times of 9.6–11.1 and 16.2–17.2 min were obtained for medium to high density and overconfluent cultures, respectively. In general, overconfluent cultures detached much slower, while their detachment rate was also decreased by the diluted harvesting solution. Moreover, several donors exhibited similar trends in cell detachment behaviour, with two clear outliers. Using the novel feature, measurements can be performed with an increased robustness, while the compact LFI design could pave the way for in situ monitoring in a variety of culture vessels, including bioreactors.

ANALYSIS OF THE POSSIBILITY OF USING THE SINGULAR VALUE DECOMPOSITION IN IMAGE COMPRESSION

In today’s highly computerized world, data compression is a key issue to minimize the costs associated with data storage and transfer. In 2019, more than 70% of the data sent over the network were images. This paper analyses the feasibility of using the SVD algorithm in image compression and shows that it improves the efficiency of JPEG and JPEG2000 compression. Image matrices were decomposed using the SVD algorithm before compression. It has also been shown that as the image dimensions increase, the fraction of eigenvalues that must be used to reconstruct the image in good quality decreases. The study was carried out on a large and diverse set of images, more than 2500 images were examined. The results were analyzed based on criteria typical for the evaluation of numerical algorithms operating on matrices and image compression: compression ratio, size of compressed file, MSE, number of bad pixels, complexity, numerical stability, easiness of implementation.

Pixel-Vernier: A General Image-Based Approach For Particle Size Distribution Estimation

Particle size distribution estimation (PSDE) is a fundamental task for heterogeneous materials characterization and modeling. This paper presents a general image-based approach for PSDE, which is called collectively “Pixel-Vernier” or PV for short. Meaningful image segmentation is the main problem to be solved for image-based PSDE. To this end, the proposed approach combines markers-controlled watershed segmentation with a clustering algorithm to solve the delineation of the boundaries of the particles. The combined approach is embedded in a coarse-to-fine strategy using a one training parameter to adapt the algorithm to the underlying distributions of the particle size. This training parameter is restricted to the size of an averaging filter. PV decomposes the image into separate particle regions. The final results of these regions are used to compute several geometric attributes for particles such as the semi-major axis, the semi-minor axis, and the equivalent diameter. Then, the geometric attributes of all the particles are used to estimate size distribution and relevant statistics. PV can be used in a laboratory as well as in a field setting. It is tested successfully on a diverse set of images that represent materials like soils, texture, rocks, and Mars surface geology.

SoloGAN: Multi-domain Multimodal Unpaired Image-to-Image Translation via a Single Generative Adversarial Network

Despite significant advances in image-to-image (I2I) translation with generative adversarial networks (GANs), it remains challenging to effectively translate an image to a set of diverse images in multiple target domains using a pair of generator and discriminator. Existing multimodal I2I translation methods adopt multiple domain-specific content encoders for different domains, where each domain-specific content encoder is trained with images from the same domain only. Nevertheless, we argue that the content (domain-invariance) features should be learned from images among all of the domains. Consequently, each domain-specific content encoder of existing schemes fails to extract the domain-invariant features efficiently. To address this issue, we present a flexible and general SoloGAN model for efficient multimodal I2I translation among multiple domains with unpaired data. In contrast to existing methods, the SoloGAN algorithm uses a single projection discriminator with an additional auxiliary classifier and shares the encoder and generator for all domains. As such, the SoloGAN model can be trained effectively with images from all domains so that the domain-invariance content representation can be efficiently extracted. Qualitative and quantitative results over a wide range of datasets against several counterparts and variants of the SoloGAN model demonstrate the merits of the method, especially for challenging I2I translation tasks, i.e., tasks that involve extreme shape variations or need to keep the complex backgrounds unchanged after translations. Furthermore, we demonstrate the contribution of each component using ablation studies.

Designing and Implementing a Method of Data Augmentation Using Machine Learning

Efficiency of neural network (NN) models depend on the parameters given and the input data. Due to the complexity of environmental conditions and limitations the data for NN models, especially for the case of images, can be insufficient. To overcome this problem data augmentation has been used to enlarge the dataset. The task is to generate diverse set of images from a small set of images for NN training. Due to data augmentation transformation, 3105 new images out of 345 input data were created for classification, detection and image segmentation.

Curve‐based crack detection using crack information gain

Crack detection provides valuable information concerning the location, extent, type, and severity for structural health monitoring of civil infrastructures. Due to the influence of uneven lighting and imaging noise, as well as the debris, various textures, and materials of the different civil structure, detecting cracks in the surface images of civil infrastructures remains challenging in that there are many false positives and defragmentation of the crack curve. In this paper, an information-based crack detection method is developed to robustly characterize and detect cracks on a curve-by-curve basis. Crack information gain (CIG) is defined to characterize the information of a local patch's being cracked. With the proposed model of estimating crack depth, the information-based crack descriptor CIG is calculated by statistically modeling the probabilistic distribution of crack depth. Secondly, crack curves of different saliency are progressively detected by iteratively searching the salient string of pixels, regardless of the small gaps between fragments of the crack curve. Finally, crack curves are validated by examining the variation of CIG along the crack curve. The experimental results on a diverse set of images of different civil infrastructures demonstrate the generalizability of the proposed method, and the overall performance on each dataset outperforms the state-of-the-art available methods with 2.9% improvement. The proposed method has potential for the quantitative evaluation of cracks on a meaningful curve basis.

Active Contour Model Using Fast Fourier Transformation for Salient Object Detection

The active contour model is a comprehensive research technique used for salient object detection. Most active contour models of saliency detection are developed in the context of natural scenes, and their role with synthetic and medical images is not well investigated. Existing active contour models perform efficiently in many complexities but facing challenges on synthetic and medical images due to the limited time like, precise automatic fitted contour and expensive initialization computational cost. Our intention is detecting automatic boundary of the object without re-initialization which further in evolution drive to extract salient object. For this, we propose a simple novel derivative of a numerical solution scheme, using fast Fourier transformation (FFT) in active contour (Snake) differential equations that has two major enhancements, namely it completely avoids the approximation of expansive spatial derivatives finite differences, and the regularization scheme can be generally extended more. Second, FFT is significantly faster compared to the traditional solution in spatial domain. Finally, this model practiced Fourier-force function to fit curves naturally and extract salient objects from the background. Compared with the state-of-the-art methods, the proposed method achieves at least a 3% increase of accuracy on three diverse set of images. Moreover, it runs very fast, and the average running time of the proposed methods is about one twelfth of the baseline.

A Novel Fusion Method for Low Brightness Enhancement Derivatives

In this paper, a straightforward and effective fusion method is designed for low brightness enhancement derivatives, which are generated through using brightness enhancement technique for a single low-brightness image. First, illumination estimation techniques and the principle of retinal imaging and cerebral cortex adjustment are combined to acquire the exposure ratio map. Then, a novel Chi-squared conversion function model and an accurate exposure ratio map are employed to obtain two derivatives with different characteristics: one is natural but not very detailed; the other is excessively bright but with prominent details. Finally, the improved weight matrix design and a novel derivatives fusion method are utilized to fuse the improved features of the derivatives. Experiments on a diverse set of images demonstrate that the proposed algorithm can not only reveal the efficiency of the brightness and detail enhancement, but also can show its superiority over several state-of-the-art processes in terms of overall visual information enhancement.

Roles of Category, Shape, and Spatial Frequency in Shaping Animal and Tool Selectivity in the Occipitotemporal Cortex.

Does the nature of representation in the category-selective regions in the occipitotemporal cortex reflect visual or conceptual properties? Previous research showed that natural variability in visual features across categories, quantified by image gist statistics, is highly correlated with the different neural responses observed in the occipitotemporal cortex. Using fMRI, we examined whether category selectivity for animals and tools would remain, when image gist statistics were comparable across categories. Critically, we investigated how category, shape, and spatial frequency may contribute to the category selectivity in the animal- and tool-selective regions. Female and male human observers viewed low- or high-passed images of round or elongated animals and tools that shared comparable gist statistics in the main experiment, and animal and tool images of naturally varied gist statistics in a separate localizer. Univariate analysis revealed robust category-selective responses for images with comparable gist statistics across categories. Successful classification for category (animals/tools), shape (round/elongated), and spatial frequency (low/high) was also observed, with highest classification accuracy for category. Representational similarity analyses further revealed that the activation patterns in the animal-selective regions were most correlated with a model that represents only animal information, whereas the activation patterns in the tool-selective regions were most correlated with a model that represents only tool information, suggesting that these regions selectively represent information of only animals or tools. Together, in addition to visual features, the distinction between animal and tool representations in the occipitotemporal cortex is likely shaped by higher-level conceptual influences such as categorization or interpretation of visual inputs.SIGNIFICANCE STATEMENT Since different categories often vary systematically in both visual and conceptual features, it remains unclear what kinds of information determine category-selective responses in the occipitotemporal cortex. To minimize the influences of low- and mid-level visual features, here we used a diverse image set of animals and tools that shared comparable gist statistics. We manipulated category (animals/tools), shape (round/elongated), and spatial frequency (low/high), and found that the representational content of the animal- and tool-selective regions is primarily determined by their preferred categories only, regardless of shape or spatial frequency. Our results show that category-selective responses in the occipitotemporal cortex are influenced by higher-level processing such as categorization or interpretation of visual inputs, and highlight the specificity in these category-selective regions.

A High-Quality VR Calibration and Real-Time Stitching Framework Using Preprocessed Features

Virtual Reality (VR) contents include $360^{\circ }\times 180^{\circ }$ seamless panoramic videos, stitched from multiple overlapping video streams. Many commercial VR devices use a two-camera rig to capture VR content. These devices have increased radial distortion effects along the seams of the stitching boundary. Moreover, a fixed number of cameras in the camera rig makes the VR system non-scalable. Since the VR experience is directly related to the quality of VR content, it is desirable to create a VR framework that is scalable in terms of the number of cameras attached to the camera-rig and has better geometric and photometric quality. In this paper, we propose an end-to-end VR system for stitching full spherical content. The VR system is composed of camera rig calibration and stitching modules. The calibration module performs a geometric alignment of the camera rig. The stitching module transforms texture from the camera or video stream into the VR stream using lookup tables (LUTs) and blend masks (BMs). In this work, our main contribution is the improvement of stitching quality. First, we propose a feature preprocessing method that filters out inconsistent, error-prone features. Secondly, we propose a geometric alignment method that outperforms state-of-the-art VR stitching solutions. We tested our system on diverse image sets and obtained state-of-the-art geometric alignment. Moreover, we achieved real-time stitching of camera and video streams up to 120 fps at 4K resolution. After stitching, we encode VR content for IP multicasting.

Jpg$Scraper$ : An Advanced Carver for JPEG Files

Orphaned file fragment carving is concerned with recovering contents of encoded data in the absence of any coding metadata. Constructing an orphaned file carver requires addressing three challenges: a specialized decoder to interpret partial file data; the ability to discriminate a specific type of encoded data from all other types of data; and comprehensive prior knowledge on possible encoding settings. In this work, we build on the ability to render a partial image contained within a segment of JPEG coded data to introduce a new carving tool that addresses all these challenges. Towards this goal, we first propose a new method that discriminates JPEG file data from among 993 file data types with 97.7% accuracy. We also introduce a method for robustly delimiting entropy coded data segments of JPEG files. This in turn allows us to identify partial JPEG file headers with zero false rejection and 0.1% of false alarm rate. Secondly, we examine a very diverse image set comprising more than 7 million images. This ensures comprehensive coverage of coding parameters used by 3,269 camera models and a wide variety of image editing tools. Further, we assess the potential impact of the developed tool on practice in terms of the amount of new evidence that it can recover. Recovery results on a set of used SD cards purchased online show that our carver is able to recover 24% more image data as compared to existing file carving tools. Evaluations performed on a standard dataset also show that Jpg Scraper improves the state-of-the-art significantly in carving JPEG file data.