Local Image Statistics Research Articles

Moving objects detection based on histogram of oriented gradient algorithm chip for hazy environment

<span>The most important aspects of computer vision are moving object detection (MOD) and tracking. Many signal-processing applications use regional image statistics. Compute-intensive video and image processing with low latency and high throughput is done with field programmable gate array (FPGA) image processing. Local image statistics are used for edge identification and filtering. The histogram of oriented gradients (HoG) algorithm extracts local shape characteristics by equalizing histograms. The objective of the work is to design the hardware chip of the algorithm and perform the simulation in the Xilinx ISE 14.7 simulation environment. The performance of the chip is evaluated in Modelsim 10.0 simulation software to check its feasibility. The performance of the chip design is estimated on Viretx-5 FPGA and compared with the MATLAB-2020 image processing tool-based response time. This form of tracking typically deals with identifying, anchoring, and tracking images and videos. A mask made from a cut-out of the object can then determine the plane's coordinates depending on its position. This type of object tracking is frequently utilized in the field of augmented reality (AR). The algorithm is most suited for object detection using hardware controllers in haze and foggy environments.</span>

Stability and Preservation of Fuzzy Membership Functions in Adaptive Gaussian Derivative Filters

This paper examines the stability and preservation of fuzzy membership functions in Gaussian filters, particularly focusing on the Adaptive Gaussian Derivative (AGD) filter. Gaussian filters are essential for smoothing and noise reduction in image processing. The AGD filter, which adapts based on local image statistics, outperforms traditional methods. Key concepts like fuzzy sets, Boundary Input Boundary Output (BIBO) stability, and convolution are defined, and the mathematical formulation of the Gaussian and its derivative is presented. The AGD filter's BIBO stability ensures bounded outputs for bounded inputs, guaranteeing consistent behavior. It also preserves fuzzy membership function properties, maintaining convexity and boundedness through linearity and continuity. Frequency response analysis using Fourier Transform confirms the AGD filter retains the Gaussian shape in the frequency domain, preserving image smoothness. Theorems and proofs validate the AGD filter's stability and its capability to preserve fuzzy membership functions, ensuring reliable processing in applications such as medical image analysis. These properties make the AGD filter a robust tool for advanced image processing tasks.

Brain MRI and CT Image Fusion Using Multiscale Local Extrema and Image Statistics

In medical applications such as radiotherapy and guided-image surgery, data fusion for diagnostic imaging has emerged as a critical issue. Because the objective of medical image fusion is to improve patient diagnosis accuracy, the fused image is created by preserving the source images' prominent details and features. It has been demonstrated that the Multi-Level Local Extrema representation has numerous advantages over traditional image modeling approaches. We propose an innovative MLE-based fusion method for multimodal medical images in this paper. In the MLE schema, the proposed algorithm decomposes source images into coarse and detailed layers, then fuses the source images using weights calculated from these detail images using image statistics. We visually and quantitatively compared the efficacy of the suggested approach to that of existing methods using five different types of medical images from various sources. The experimental results showed that the proposed scheme outperforms other current typical schemes in terms of both qualitative image quality and objective evaluation.

A blending method for forward-looking sonar mosaicing handling intra- and inter-frame artifacts

Forward-looking sonar (FLS) has been gaining attractions in the realm of near-bottom, close-range underwater inspections, owing to its high-resolution and rapid framerate capabilities. Although automatic target recognition (ATR) algorithms have been tentatively employed for object detection tasks, the necessity for human oversight remains vital, particularly in sensitive areas. A comprehensive FLS mosaic encapsulating all relevant information is highly desired to aid experts in managing an extensive array of perception data. Yet, prior research has often presumed that FLS operates under an ideal system configuration, assuming optimal sonar imaging setups and the availability of precise positioning data. Without these assumptions, intra-frame and inter-frame artifacts would emerge, deteriorating the quality of the resultant mosaic by obscuring essential information. In this study, we propose an innovative blending method specifically for FLS mosaicing aiming preserve important information. We first develop a long–short time sliding window (LST-SW) to rectify the local statistics of raw sonar images. These statistics then serve to create a global variance map (GVM). This step helps to emphasize the useful information contained in images in the blending phase by classifying the informative and featureless pixels, thereby enhancing the quality of final mosaic. This approach is substantiated through data collected in real-world scenarios. The results show that our method can preserve more details in FLS mosaics for human inspection purposes in practice.

Jointly learning and training: using style diversification to improve domain generalization for deepfake detection

<abstract><p>Most existing deepfake detection methods often fail to maintain their performance when confronting new test domains. To address this issue, we propose a generalizable deepfake detection system to implement style diversification by alternately learning the domain generalization (DG)-based detector and the stylized fake face synthesizer (SFFS). For the DG-based detector, we first adopt instance normalization- and batch normalization-based structures to extract the local and global image statistics as the style and content features, which are then leveraged to obtain the more diverse feature space. Subsequently, contrastive learning is used to emphasize common style features while suppressing domain-specific ones, and adversarial learning is performed to obtain the domain-invariant features. These optimized features help the DG-based detector to learn generalized classification features and also encourage the SFFS to simulate possibly unseen domain data. In return, the samples generated by the SFFS would contribute to the detector's learning of more generalized features from augmented training data. Such a joint learning and training process enhances both the detector's and the synthesizer's feature representation capability for generalizable deepfake detection. Experimental results demonstrate that our method outperforms the state-of-the-art competitors not only in intra-domain tests but particularly in cross-domain tests.</p></abstract>

Local image statistics can account for the perceived naturalness of image contrast

Local image statistics can account for the perceived naturalness of image contrast

Region-Growing-Based Automatic Localized Adaptive Thresholding Algorithm for Water Extraction Using Sentinel-2 MSI Imagery

Water is a distinct land cover feature on the earth. Water can easily be extracted from satellite data under known/controlled conditions using simple thresholding techniques. However, these pixel thresholding techniques fail when applied over large regions because they cannot adapt the thresholds based on local variability. Although region-growing methods are available to overcome this issue, there is a limitation in using a global threshold value determined for a satellite scene. Such thresholds fail to adapt to the variations within the scene. Hence, this study proposed a new region-growing-based approach for improved classification of water features using Sentinel-2 Multi-Spectral Instrument (MSI) data. The significant difference between this approach and other region-growing methods is that this technique uses dynamic thresholding for localized adaption. The local image statistics of the Normalized Different Water Index (NDWI) layer are used for adding new pixels to the region, and the threshold is adjusted for every new pixel added. This technique also minimizes manual intervention in water classification, thus enabling total process automation in near-real-time. The thresholds are dynamically determined not only for each scene but also for each new pixel being classified for the accurate delineation of the water pixels. A rigorous quality evaluation was done on the resultant water layer. This technique achieved an overall accuracy of 84.04% in challenging conditions.

Texture and material classification with multi-scale ternary and septenary patterns

This paper ideates, inspired by LBP and its variants, a novel local feature extraction operator for texture classification, referred to as Multi-scale Ternary and Septenary Pattern (MTSP). MTSP is a histogram-based feature reṕresentation that is composed of two single-scale STP and SSP (single-scale ternary and septenary patterns, respectively) encoders designed according to a novel set theory based pattern encoding scheme that integrates the concepts of both LQP’s and LTP’s operators. The essence of STP and SSP is to compute several virtual pixels based on various local and global image statistics and progressively encode interactions between local and non-local pixels by examining the directional information and differential excitation according to relationships between adjacent pixels rearranged in a variety of spatial arrangements. Unlike various parametric state-of-the-art texture operators that perform thresholding based on static thresholds, MTSP incorporates dynamic thresholds estimated automatically. MTSP descriptor has good ability as faithfully as possible to capture more detailed image information via complementary texture information generated from the fusion of both STP and SSP encoders. Experimental results show that MTSP ensures reliable performance stability over ten texture datasets and against several recent representative methods. In addition, the performance of MTSP is further proved statistically via the Wilcoxon signed rank test demonstrating thus that MTSP is a good candidate for texture modeling.

High-order local image statistics are used in radiographic judgments of breast density

High-order local image statistics are used in radiographic judgments of breast density

Functional recursion of orientation cues in figure-ground separation

Visual texture is an important cue to figure-ground organization. While processing of texture differences is a prerequisite for the use of this cue to extract figure-ground organization, these stages are distinct processes. One potential indicator of this distinction is the possibility that texture statistics play a different role in the figure vs. in the ground. To determine whether this is the case, we probed figure-ground processing with a family of local image statistics that specified textures that varied in the strength and spatial scale of structure, and the extent to which features are oriented. For image statistics that generated approximately isotropic textures, the threshold for identification of figure-ground structure was determined by the difference in correlation strength in figure vs. ground, independent of whether the correlations were present in figure, ground, or both. However, for image statistics with strong orientation content, thresholds were up to two times higher for correlations in the ground, vs. the figure. This held equally for texture-defined objects with convex or concave boundaries, indicating that these threshold differences are driven by border ownership, not boundary shape. Similar threshold differences were found for presentation times ranging from 125 to 500 ms. These findings identify a qualitative difference in how texture is used for figure-ground analysis, vs. texture discrimination. Additionally, it reveals a functional recursion: texture differences are needed to identify tentative boundaries and consequent scene organization into figure and ground, but then scene organization modifies sensitivity to texture differences according to the figure-ground assignment.

Random-Valued Impulse Noise Detection and Removal based on Local Statistics of Images

Random-valued impulse noise removal from images is a challenging task in the field of image processing and computer vision. In this paper, an effective three-step noise removal method was proposed using local statistics of grayscale images. Unlike most existing denoising algorithms that assume the noise density is known, our method estimated the noise density in the first step. Based on the estimated noise density, a noise detector was implemented to detect corrupted pixels in the second step. Finally, a modified weighted mean filter was utilized to restore the detected noisy pixels while leaving the noise-free pixels unchanged. The noise removal performance of our method was compared with 10 well-known denoising algorithms. Experimental results demonstrated that our proposed method outperformed other denoising algorithms in terms of noise detection and image restoration in the vast majority of the cases.

Noise estimation in single coil MR images

Magnetic resonance (MR) signal is degraded by acquisition noise. Estimation and removal of noise from MR images is essential for analysis and further pre-processing. The work proposes a new method for noise variance estimation in single coil magnitude MR images. The proposed noise estimator is derived using the Maximum Á Posterior (MAP) estimator. Noise follows a Gaussian distribution at high Signal to Noise (SNR) ratio and Rician distribution at low SNR. In the MAP framework, Gaussian noise model and local statistics of the noisy image are instrumental in estimation of noise at high SNR. The proposed work also exploits the local second order moments in the MAP framework for noise estimation at low SNR. Convergence of the proposed MAP framework further highlights the optimal performance of the noise estimator. Noise estimation experiments conducted on Brainweb database exhibit encouraging performance at high as well as low SNR.

Redundancy between spectral and higher-order texture statistics for natural image segmentation

Visual texture, defined by local image statistics, provides important information to the human visual system for perceptual segmentation. Second-order or spectral statistics (equivalent to the Fourier power spectrum) are a well-studied segmentation cue. However, the role of higher-order statistics (HOS) in segmentation remains unclear, particularly for natural images. Recent experiments indicate that, in peripheral vision, the HOS of the widely adopted Portilla-Simoncelli texture model are a weak segmentation cue compared to spectral statistics, despite the fact that both are necessary to explain other perceptual phenomena and to support high-quality texture synthesis. Here we test whether this discrepancy reflects a property of natural image statistics. First, we observe that differences in spectral statistics across segments of natural images are redundant with differences in HOS. Second, using linear and nonlinear classifiers, we show that each set of statistics individually affords high performance in natural scenes and texture segmentation tasks, but combining spectral statistics and HOS produces relatively small improvements. Third, we find that HOS improve segmentation for a subset of images, although these images are difficult to identify. We also find that different subsets of HOS improve segmentation to a different extent, in agreement with previous physiological and perceptual work. These results show that the HOS add modestly to spectral statistics for natural image segmentation. We speculate that tuning to natural image statistics under resource constraints could explain the weak contribution of HOS to perceptual segmentation in human peripheral vision.

Gravitational effects of scene information in object localization

We effortlessly interact with objects in our environment, but how do we know where something is? An object’s apparent position does not simply correspond to its retinotopic location but is influenced by its surrounding context. In the natural environment, this context is highly complex, and little is known about how visual information in a scene influences the apparent location of the objects within it. We measured the influence of local image statistics (luminance, edges, object boundaries, and saliency) on the reported location of a brief target superimposed on images of natural scenes. For each image statistic, we calculated the difference between the image value at the physical center of the target and the value at its reported center, using observers’ cursor responses, and averaged the resulting values across all trials. To isolate image-specific effects, difference scores were compared to a randomly-permuted null distribution that accounted for any response biases. The observed difference scores indicated that responses were significantly biased toward darker regions, luminance edges, object boundaries, and areas of high saliency, with relatively low shared variance among these measures. In addition, we show that the same image statistics were associated with observers’ saccade errors, despite large differences in response time, and that some effects persisted when high-level scene processing was disrupted by 180° rotations and color negatives of the originals. Together, these results provide evidence for landmark effects within natural images, in which feature location reports are pulled toward low- and high-level informative content in the scene.

An effective feature extraction method for rice leaf disease classification

Our society is getting more and more technology dependent day by day. Nevertheless, agriculture is imperative for our survival. Rice is one of the primary food grains. It provides sustenance to almost fifty percent of the world population and promotes huge amount of employments. Hence, proper mitigation of rice plant diseases is of paramount importance. A model to detect three rice leaf diseases, namely bacterial leaf blight, brown spot, and leaf smut is proposed in this paper. Backgrounds of the images are removed by saturation threshold while disease affected areas are segmented using hue threshold. Distinctive features from color, shape, and texture domain are extracted from affected areas. These features can robustly describe local and global statistics of such images. Trying a couple of classification algorithms, extreme gradient boosting decision tree ensemble is incorporated in this model for its superior performance. Our model achieves 86.58% accuracy on rice leaf diseases dataset from UCI, which is higher than previous works on the same dataset. Class-wise accuracy of the model is also consistent among the classes.

A statistical active contour model for interactive clutter image segmentation using graph cut optimization

This paper presents a statistical region based Active Contour Model (ACM) considering the correlation between local and global image statistics to segment cluttered images. Generally, cluttered images do not have constant intensity distribution; rather, the intensity may follow near constant variation in different regions. To quantify this variation, we have considered the Coefficient of Variation (CoV) of the regions interior and exterior to the contour as global statistics and the CoV in the local patches as local statistics. Subsequently, the region energy term of the proposed ACM is designed such that it minimizes the difference between the local and global statistics i.e. it encourages CoV for all the local patches inside and outside of the final contour to be nearly homogeneous. Further, we have verified that the energy formulation can be efficiently discretized and solved using graph cut optimization. The main advantages of graph-based formulation over level set formulation are the existence of a global optimal solution and lesser sensitivity to contour initialization. Additionally, the former formulation is significantly faster being non-iterative or convergable with very few iterations. Experimental results demonstrate the superior performance of our approach against other state-of-the-art active contour approaches and also over its level set counterpart.

Visible-Infrared Image Fusion Based on Early Visual Information Processing Mechanisms

In this work, we simulate the early visual information processing mechanisms in biological visual system (BVS) to solve the visible-infrared image fusion (VIIF) task. Concretely, both infrared and visible images are first processed with a dynamic receptive field (DRF), which is imitated by a Difference of Gaussian (DoG) function whose parameters are adjusted according to local image statistics (e.g., local edge responses). The DRF processing produces two components for each source image (e.g., the visible image or the infrared image) that respectively represent the results of On-center based DRF and Off-center based DRF. Then, the results of On-center based DRF for visible image are fused with the results of Off-center based DRF for infrared image and the results of On-center based DRF for infrared image are fused with the results of Off-center based DRF for visible image, according to the mechanisms of cortex-based center-surround fusion. Algorithmically, this step fuses the visible and infrared images processed by DRF with On-center and Off-center according to the different levels of local homogeneity. Moreover, a feedward signal acting as the sub-cortical flow is also introduced to adjust the results during the fusion. The final output image is simply obtained by the summation of two components after the cortex and sub-cortex based fusion. Qualitative and quantitative tests on four datasets demonstrate that our approach can fuse the infrared and visible images effectively with the good background details and discernible salient areas. We emphasize the importance of corticothalamic feedback, cortex and sub-cortex-based fusion, and the interactions between On-center pathway and Off-center pathway that are ubiquitous in BVS for producing the high-quality visual signals.

When Scenes Look Like Materials: René Magritte’s Reversible Figure–Ground Motif

We draw attention to a frequent motif in the work of the Belgian surrealist René Magritte (1898–1967). In the motif, a scene is depicted that contains a silhouette, which itself contains another depicted scene. The silhouette is bistable, appearing either as a figural region whose positive space is covered, or filled, with the interior scene texture, or as a ground region providing a window onto a more distant scene. We call this the ‘reversible figure–ground motif’. Because the stimulus does not change when our percept changes, the motif’s appearance at any particular moment cannot be explained by its local or global image statistics. Instead principles of perceptual organization, and in particular image segmentation and figure–ground assignment, appear crucial for determining whether the interior of the silhouette is processed as a material vs. a scene — which in turn reflects the fundamental role of visual segmentation in material and scene perception more generally.

Segmentation‐based MAP despeckling of medical ultrasound images in shearlet domain based on normal inverse Gaussian distribution

Speckle noise is an undesirable phenomenon that exhibits granular pattern which reduces the diagnostic capability of clinical ultrasound (US) images. In this study, a new speckle denoising method based on modelling of detail band shearlet coefficients of log-transformed US images is presented. In each detail shearlet subband, coefficients corresponding to signal and speckle noise are modelled as normal inverse Gaussian and Gaussian priors, respectively. These coefficients, based on image local statistics are segmented into different regions of heterogeneity viz. homogeneous, heterogeneous and strongly heterogeneous regions, respectively, so as to control over smoothing of denoised images. Then, using the prior distributions maximum a posteriori (MAP) estimation is performed over all regions of detail bands except those regions that represent strongly heterogeneous coefficients. For better performance, an adaptive weight function is also used in the MAP expression which reduces the loss of feature information. Experimentation is done on noise-free synthetic kidney and foetus US images and a set of 60 real US images. The results are presented for objective and subjective quality assessment of the proposed method and five other methods for speckle denoising. The potential of the proposed method in comparison to other methods can easily be ascertained from the obtained denoising results.

Speckle and Rician Noise Removal from Medical Images and Ultrasound Images

In medical images, medical images are corrupted by different types of noise. It is important to get a precise picture and accurately observe the correspondence. Removing noise from medical images has become a very difficult problem in the field of the medical image. The most well-known noise reduction method, which is usually based on the local statistics of medical images, is efficient because of the noise reduction of medical images. In paper, an efficient and simple method for noise reduction from medical images is presented. The paper proposes a filtering system to combine both the Median filter and Gaussian filter to remove the Speckle noise form Medical and Ultrasound images. The image quality is measured through statistical quantities: Peak signal to noise ratio (PSNR). Experimental results show that the proposed system removes Speckle noise from medical images.