Explore the vast range of titles available.

Animal colour patterns are a model system for understanding evolution because they are unusually accessible for study and experimental manipulation. This is possible because their functions are readily identifiable. In this final paper of the symposium we provide a diagram of the processes affecting colour patterns and use this to summarize their functions and put the other papers in a broad context. This allows us to identify significant ‘holes’ in the field that only become obvious when we see the processes affecting colour patterns, and their interactions, as a whole. We make suggestions about new directions of research that will enhance our understanding of both the evolution of colour patterns and visual signalling but also illuminate how the evolution of multiple interacting traits works. This article is part of the themed issue ‘Animal coloration: production, perception, function and application’.

Haralick texture features are common texture descriptors in image analysis. To compute the Haralick features, the image gray-levels are reduced, a process called quantization. The resulting features depend heavily on the quantization step, so Haralick features are not reproducible unless the same quantization is performed. The aim of this work was to develop Haralick features that are invariant to the number of quantization gray-levels. By redefining the gray-level co-occurrence matrix (GLCM) as a discretized probability density function, it becomes asymptotically invariant to the quantization. The invariant and original features were compared using logistic regression classification to separate two classes based on the texture features. Classifiers trained on the invariant features showed higher accuracies, and had similar performance when training and test images had very different quantizations. In conclusion, using the invariant Haralick features, an image pattern will give the same texture feature values independent of image quantization.

Modelling the mapping from scene irradiance to image intensity is essential for many computer vision tasks. Such mapping is known as the camera response. Most digital cameras use a nonlinear function to map irradiance, as measured by the sensor to an image intensity used to record the photograph. Modelling of the response is necessary for the nonlinear calibration. In this paper, a new high-performance camera response model that uses a single latent variable and fully connected neural network is proposed. The model is produced using unsupervised learning with an autoencoder on real-world (example) camera responses. Neural architecture searching is then used to find the optimal neural network architecture. A latent distribution learning approach was introduced to constrain the latent distribution. The proposed model achieved state-of-the-art CRF representation accuracy in a number of benchmark tests, but is over twice as fast as the best current models when performing the maximum likelihood estimation during camera response calibration due to the simple yet efficient model representation.

Low-lighting is a common condition in data collection due to environmental restrictions. However, high-level pattern recognition tasks such as object detection require the datasets to be more clear. Thus, low-light image enhancement is necessary. Noise and color distortion are two major problems of the existing enhancement algorithms. This paper has proposed a low-light image enhancement algorithm that integrates denoising and color restoration. First, we propose a two-stage hybrid decomposition network, which can perform modified Retinex-decomposition on paired images, and then extract principal components of the decomposed low-light images to handle the nonlinear residuals, thereby obtaining reliable reflectance and illumination maps. Then, in order not to over-smooth the details and edges of the image, we use a flexible joint function to train the hybrid network. Finally, we create a color regulator in the HSI (Hue-Saturation-Intensity) space to correct the distortion in RGB space caused by coupling between pixels. Experimental results on public datasets show that the proposed method greatly enhanced the quality of low-light images.

Color face recognition has more attention recently since it considered one of the most popular biometric pattern recognitions. With a considerable development in multimedia technologies, finding a suitable color information extraction from color images becomes a hard problem. Several color face recognition methods have been developed. However, these methods still suffer from some limitations, such as increasing the number of extracted features, which leads to an increase in computational time. Besides, among those features some of them are redundant and irrelevant that will influence the quality of the recognition. Therefore, this paper presents a novel color face recognition method that depends on a new family of fractional-order orthogonal functions, which is called orthogonal fractional-order exponent functions. Then, using these functions as the basis functions of novel multi-channel orthogonal fractional-order exponent moments (FrMEMs), these novel descriptors are defined in polar coordinates over the unit circle and have many characteristics. A set of experimental series are performed using a set of well-known color face recognition and compared with other CFR techniques. Besides, a group of feature selection methods with different classifiers used to evaluate the number of extracted features is suitable or needs to be enhanced. Experimental results illustrate that the proposed method based on FrMEMs outperforms other CFR methods. As well as, the recognition rate doesn’t influence by reducing the number of features using different FS methods.

Robotic fabric manipulation has applications in home robotics, textiles, senior care and surgery. Existing fabric manipulation techniques, however, are designed for specific tasks, making it difficult to generalize across different but related tasks. We build upon the Visual Foresight framework to learn fabric dynamics that can be efficiently reused to accomplish different sequential fabric manipulation tasks with a single goal-conditioned policy. We extend our earlier work on VisuoSpatial Foresight (VSF), which learns visual dynamics on domain randomized RGB images and depth maps simultaneously and completely in simulation. In this earlier work, we evaluated VSF on multi-step fabric smoothing and folding tasks against 5 baseline methods in simulation and on the da Vinci Research Kit surgical robot without any demonstrations at train or test time. A key finding was that depth sensing significantly improves performance: RGBD data yields an 80% improvement in fabric folding success rate in simulation over pure RGB data. In this work, we vary 4 components of VSF, including data generation, visual dynamics model, cost function, and optimization procedure. Results suggest that training visual dynamics models using longer, corner-based actions can improve the efficiency of fabric folding by 76% and enable a physical sequential fabric folding task that VSF could not previously perform with 90% reliability. Code, data, videos, and supplementary material are available at https://sites.google.com/view/fabric-vsf/.

Conflict-induced control refers to humans’ ability to regulate attention in the processing of target information (e.g., the color of a word in the color-word Stroop task) based on experience with conflict created by distracting information (e.g., an incongruent color word), and to do so either in a proactive (preparatory) or a reactive (stimulus-driven) fashion. Interest in conflict-induced control has grown recently, as has the awareness that effects attributed to those processes might be affected by conflict-unrelated processes (e.g., the learning of stimulus-response associations). This awareness has resulted in the recommendation to move away from traditional interference paradigms with small stimulus/response sets and towards paradigms with larger sets (at least four targets, distractors, and responses), paradigms that allow better control of non-conflict processes. Using larger sets, however, is not always feasible. Doing so in the Stroop task, for example, would require either multiple arbitrary responses that are difficult for participants to learn (e.g., manual responses to colors) or non-arbitrary responses that can be difficult for researchers to collect (e.g., vocal responses in online experiments). Here, we present a spatial version of the Stroop task that solves many of those problems. In this task, participants respond to one of six directions indicated by an arrow, each requiring a specific, non-arbitrary manual response, while ignoring the location where the arrow is displayed. We illustrate the usefulness of this task by showing the results of two experiments in which evidence for proactive and reactive control was obtained while controlling for the impact of non-conflict processes.

This paper presents a novel medical image encryption scheme that integrates a synchronously updated chaotic logical Boolean network with the Four-Color Theorem to achieve high security and structural obfuscation. The proposed Boolean network, constructed through the semi-tensor product and derived from the Hénon map, exhibits enhanced dynamical properties, such as increased sensitivity to initial conditions and stronger chaotic behavior, thereby improving cryptographic unpredictability and robustness. To preserve clinically significant image information, a color-labeling strategy is employed to identify and encode diagnostically relevant regions within the image. A color label matrix, generated according to the Ffour-Color Theorem and matched to the dimensions of the plaintext image, is subsequently employed to guide pixel position scrambling. This process effectively conceals anatomical and pathological features while maintaining computational efficiency. Experimental results confirm the robustness of the proposed scheme, demonstrating strong resistance against statistical and differential attacks.

We describe a novel probabilistic framework for real-time tracking of multiple objects from combined depth-colour imagery. Object shape is represented implicitly using 3D signed distance functions. Probabilistic generative models based on these functions are developed to account for the observed RGB-D imagery, and tracking is posed as a maximum a posteriori problem. We present first a method suited to tracking a single rigid 3D object, and then generalise this to multiple objects by combining distance functions into a shape union in the frame of the camera. This second model accounts for similarity and proximity between objects, and leads to robust real-time tracking without recourse to bolt-on or ad-hoc collision detection.

Simultaneous illumination and communication using solid-state lighting devices like white light-emitting diode (LED) light sources is gaining popularity. The white light LED comprises a single-colored yellow phosphor excited by the blue LED chip. Therefore, color-quality determining parameters like color-rendering index (CRI), correlated color temperature (CCT), and CIE 1931 chromaticity coordinates of generic white LED sources are poor. This article presents the development of multi-color phosphors excited by a blue LED to improve light quality and bandwidth. A multi-layer stacking of phosphor layers excited by a blue LED led to the quenching of photoluminescence (PL) and showed limited bandwidth. To solve this problem, a lens-free, electrically powered, broadband white light source is designed by mounting multi-color phosphor LEDs in a co-planar ring-topology. The CRI, CCT, and CIE 1931 chromaticity coordinates of the designed lamp (DL) were found to be 90, 5114 K, and (0.33, 0.33), respectively, which is a good quality lamp for indoor lighting. CRI of DL was found to be 16% better than that of white LED (WL). Assessment of visible light communications (VLC) feasibility using the DL includes time interval error (TIE) of data pattern or jitter analysis, eye diagram, signal-to-noise ratio (SNR), fast Fourier transform (FFT), and power spectral density (PSD). DL transmits binary data stream faster than WL due to a reduction in rise time and total jitter by 31% and 39%, respectively. The autocorrelation function displayed a narrow temporal pulse for DL. The DL is beneficial for providing high-quality illumination indoors while minimizing PL quenching. Additionally, it is suitable for indoor VLC applications.