Image Filtering Research Articles

The Subversive AI Acceptance Scale (SAIA-8): A Scale to Measure User Acceptance of AI-Generated, Privacy-Enhancing Image Modifications

To resist government and corporate use of facial recognition to surveil users through their personal images, researchers have created privacy-enhancing image filters that use adversarial machine learning. These "subversive AI'' (SAI) image filters aim to defend users from facial recognition by distorting personal images in ways that are barely noticeable to humans but confusing to computer vision algorithms. SAI filters are limited, however, by the lack of rigorous user evaluation that assess their acceptability. We addressed this limitation by creating and validating a scale to measure user acceptance --- the SAIA-8. In a three-step process, we apply a mixed-methods approach that closely adhered to best practices for scale creation and validation in measurement theory. Initially, to understand the factors that influence user acceptance of SAI filter outputs, we interviewed 15 participants. Interviewees disliked extant SAI filter outputs because of a perceived lack of usefulness and conflicts with their desired self-presentation. Using insights and statements from the interviews, we generated 106 potential items for the scale. Employing an iterative refinement and validation process with 245 participants from Prolific, we arrived at the SAIA-8 scale: a set of eight items that capture user acceptability of privacy-enhancing perturbations to personal images, and that can aid in benchmarking and prioritizing user acceptability when developing and evaluating new SAI filters.

Integral field spectroscopy supports atmospheric optics to reveal the finite outer scale of the turbulence

The spatial coherence wavefront outer scale ( characterizes the size of the largest turbulence eddies in Earth's atmosphere, determining low spatial frequency perturbations in the wavefront of the light captured by ground-based telescopes. Advances in adaptive optics (AO) techniques designed to compensate for atmospheric turbulence emphasize the crucial role of this parameter for the next generation of large telescopes. The motivation of this work is to introduce a novel technique for estimating \ from seeing-limited integral field spectroscopic (IFS) data. This approach is based on the impact of a finite \ on the light collected by the pupil entrance of a ground-based telescope. We take advantage of the homogeneity of IFS observations to generate band filter images spanning a wide wavelength range, enabling the assessment of image quality (IQ) at the telescope's focal plane. Comparing the measured wavelength-dependent IQ variation with predictions derived from a first-order analytical approach based on turbulence statistics simplifications using the von Kármán model provides valuable insights into the prevailing \ parameter during the observations. We applied the proposed technique to observations from the Multi-Unit Spectroscopic Explorer (MUSE) in the wide-field mode obtained at the Paranal Observatory. Our analysis successfully validates the first-order analytical expression, which combines the seeing ($ $) and the \ parameters, to predict the IQ variations with the wavelength in ground-based astronomical data. However, we observed some discrepancies between the measured and predictions of the IQ that are analyzed in terms of uncertainties in the estimated $ $ and dome-induced turbulence contributions. This work constitutes the empirical validation of the analytical expression for estimating IQ at the focal plane of ground-based telescopes under specific $ $ and finite \ conditions. Additionally, we provide a simple methodology to characterize the \ and dome-seeing ($ dome $) as by-products of IFS observations routinely conducted at major ground-based astronomical observatories.

Assessment of SPECT Image Reconstruction in Liver Scanning Using 99mTc/ EDDA/ HYNIC-TOC

Purpose: Given that the Single Photon Emission Computed Tomography (SPECT) image quality is defined experimentally, developing a specialized scanning technique for each procedure is necessary to increase the diagnosis accuracy. This study aims to determine the optimal algorithm for liver scan reconstruction using 99mTc/SPECT. Materials and Methods: The Filtered Back-Projection (FBP) reconstruction method was used in liver scanning using 99mTc-EDDA/HYNIC-TOC (Tektrotyd) for SPECT images of 30 patients which were acquired with a dual-head EvoExel detector system. Using different types of filters in SPECT imaging, various optimal results can be achieved in the processed images, such as artifact reduction, noise reduction, or signal enhancement and recovery. To evaluate the effect of different filters on image quality, Signal-to-Noise-Ratio (SNR), Contrast-to-Noise-Ratio (CNR), and contrast parameters were calculated. Results: Applying filters enhanced contrast in the images in most cases as well as CNR and SNR. Metz (power = 2), Shepp-Logan (Cut-off frequency = 0.67) and Metz (power = 2) filters increase the CNR, contrast and SNR in images more than the other filters, respectively. The maximum improvement for CNR, contrast and SNR was from 0.62 to 2.35, 0.99 to 1.31, and 8.48 to 14.70, respectively. Conclusion: Based on the results, the Hamming filter, due to providing high-quality images for visual analysis of liver SPECT, and the Butterworth filter, due to balancing the image quality and noise for quantitative analysis, are recommended.

Automatic classification of spread‐F types in ionogram images using support vector machine and convolutional neural network

AbstractAn ionogram image serves as a valuable data for examining the ionospheric bottom side characteristics and variabilities. Spread-F is indicated or identified by plasma irregularity in the ionospheric region. Diffused echo in the ionogram images particularly pose challenges for efficient interpretation required in further applications. An automatic classification of spread-F is presented in this study. Ionogram images are automatically classified using preprocessing techniques to improve the classification performance. In this study, the classification is designed by two machine learning algorithms, including support vector machine (SVM) and convolutional neural network (CNN). The CNN model with preprocessing technique outperforms the SVM alternative based on 4,692 labelled ionogram images from the FMCW-type ionosonde at Chumphon station, Thailand. The model successfully classified clear, frequency spread-F (FSF), range spread-F (RSF), strong spread-F (SSF), and unidentified class with an accuracy of 98.0%, 85.1%, 90.7%, 66.7%, and 99.2%, respectively. The proposed automatic classification models achieved to classify classes of ionogram images. In addition, the image filtering and data preprocessing are useful with ionogram images for improving the model classification performance. Graphical Abstract

Exploring AI-driven Innovations in Image Communication Systems for Enhanced Medical Imaging Applications

Artificial intelligence (AI) has emerged as a promising avenue for enhancing medical imaging systems and improving clinical workflows. This research explores innovative applications of AI and deep learning for image communication networks in healthcare. Specifically, we develop an intelligent image compression framework that optimizes data transmission and speeds interpretation of radiology scans. Our approach combines convolutional neural networks, generative adversarial networks, and specialized image filters to balance communication efficiency, diagnostic accuracy, and system latency. Rigorous experiments validate superior performance over traditional methods and commercial products across modalities including MRI, CT, and ultrasound. Crucially, the proposed methods demonstrate expert-level precision in anatomy labeling and pathology detection. By intelligently streamlining image transfer and analytics, this AI-powered system could facilitate ubiquitous, real-time diagnostics via telemedicine. Enhanced connectivity between imaging devices and clinical specialists can improve patient outcomes and reduce healthcare costs. Our solutions set the stage for more advanced AI integration in imaging networks and data-intensive medicine

Imaging detection of the inner dust belt and the four exoplanets in the HR 8799 system with JWST's MIRI coronagraph

The MIRI instrument on board JWST is now offering high-contrast imaging capacity at mid-IR wavelengths, thereby opening a completely new field of investigation for characterizing young exoplanetary systems. The multiplanet system HR\,8799 is the first target observed with MIRI's coronagraph as part of the MIRI-EC Guaranteed Time Observations (GTO) exoplanet program, launched in November 2022. We obtained deep observations in three coronagraphic filters, from $ 10$ to 15\,muup m (F1065C, F1140C, F1550C), and one standard imaging filter at $ (F2100W). The goal of this work is to extract photometry for the four planets and to detect and investigate the distribution of circumstellar dust. Using dedicated observations of a reference star, we tested several algorithms to subtract the stellar diffraction pattern, while preserving the fluxes of planets, which can be significantly affected by over-subtraction. To obtain correct measurements of the planet's flux values, the attenuation by the coronagraphs as a function of their position must be accounted for, as well as an estimation of the normalisation with respect to the central star. We tested several procedures to derive averaged photometric values and error bars. These observations have enabled us to obtain two main results. First, the four planets in the system are well recovered and we were able to compare their mid-IR fluxes, combined with near-IR flux values from the literature, to two exoplanet atmosphere models: ATMO and Exo-REM . As a main outcome, the MIRI photometric data points imply larger radii ( 1.04 or 1.17\,$R_ J for planet b) and cooler temperatures ( 950 or 1000\,K for planet b), especially for planet b, in better agreement with evolutionary models. Second, these JWST/MIRI coronagraphic data also deliver the first spatially resolved detection of the inner warm debris disk, the radius of which is constrained to about 15\,au, with flux densities that are comparable to (but lower than) former unresolved spectroscopic measurements with Spitzer. The coronagraphs coming from MIRI ushers in a new vision of known exoplanetary systems that differs significantly from shorter wavelength, high-contrast images delivered by extreme adaptive optics from the ground. Inner dust belts and background galaxies become dominant at some mid-IR wavelengths, potentially causing confusion in detecting exoplanets. Future observing strategies and data reductions ought to take such features into account.

Infrared and Visual Image Fusion Based on a Local-Extrema-Driven Image Filter.

The objective of infrared and visual image fusion is to amalgamate the salient and complementary features of the infrared and visual images into a singular informative image. To accomplish this, we introduce a novel local-extrema-driven image filter designed to effectively smooth images by reconstructing pixel intensities based on their local extrema. This filter is iteratively applied to the input infrared and visual images, extracting multiple scales of bright and dark feature maps from the differences between continuously filtered images. Subsequently, the bright and dark feature maps of the infrared and visual images at each scale are fused using elementwise-maximum and elementwise-minimum strategies, respectively. The two base images, representing the final-scale smoothed images of the infrared and visual images, are fused using a novel structural similarity- and intensity-based strategy. Finally, our fusion image can be straightforwardly produced by combining the fused bright feature map, dark feature map, and base image together. Rigorous experimentation conducted on the widely used TNO dataset underscores the superiority of our method in fusing infrared and visual images. Our approach consistently performs on par or surpasses eleven state-of-the-art image-fusion methods, showcasing compelling results in both qualitative and quantitative assessments.

Supernova environments in J-PLUS

We investigated the local environmental properties of 418 supernovae (SNe) of all types using data from the Javalambre Photometric Local Universe Survey (J-PLUS), which includes five broad-band and seven narrow-band imaging filters. Our study involves two independent analyses: (1) the normalized cumulative-rank (NCR) method, which utilizes all 12 single bands along with five continuum-subtracted narrow-band emission and absorption bands, and (2) simple stellar population (SSP) synthesis, where we build spectral energy distributions (SED) of the surrounding 1 kpc2 SN environment using the 12 broad- and narrow-band filters. Improvements on previous works include: (i) the extension of the NCR technique to other filters (broad and narrow) and the use a set of homogeneous data (same telescope and instruments); (ii) a correction for extinction to all bands based on the relation between the g − i color and the color excess E(B − V); and (iii) a correction for the contamination of the [N II] λ6583 line that falls within the Hα filter. All NCR distributions in the broad-band filters, tracing the overall light distribution in each galaxy, are similar to each other. The main difference is that type Ia, II, and IIb SNe are preferably located in redder environments than the other SN types. The radial distribution of the SNe shows that type IIb SNe seem to have a preference for occurring in the inner regions of galaxies, whereas other types of SNe occur throughout the galaxies without a distinct preference for a specific location. For the Hα filter we recover the sequence from SNe Ic, which has the highest NCR, to SNe Ia, which has the lowest; this is interpreted as a sequence in progenitor mass and age. All core-collapse SN types are strongly correlated to the [O II] emission, which also traces star formation rate (SFR), following the same sequence as in Hα. The NCR distributions of the Ca II triplet show a clear division between II-IIb-Ia and Ib-Ic-IIn subtypes, which is interpreted as a difference in the environmental metallicity. Regarding the SSP synthesis, we found that including the seven J-PLUS narrow filters in the fitting process has a more significant effect on the core-collapse SN environmental parameters than for SNe Ia, shifting their values toward more extincted, younger, and more star-forming environments, due to the presence of strong emission lines and stellar absorptions in those narrow bands.

Web Framework Content Based Filtering for Picture Sharing

Information technology has greatly impacted the multimedia industry, particularly in the areas of video and audio content. There are various applications available for managing individual videos and converting them into formats suitable for online sharing. With the widespread use of mobile phones and computers, capturing and storing pictures and videos has become a common practice. These files can be easily shared on different networks, including popular platforms like Facebook, YouTube, and WhatsApp. The rise of web 2.0 has led to an increase in file sharing, including explicit content, as uploads are often not properly filtered. Social media platforms play a significant role in facilitating real-time video sharing, but they also host unwanted content such as violent and pornographic videos. This research aims to develop a web framework application that filters images and videos before they are uploaded to interconnected networks, ensuring that unwanted content is detected and flagged. Users will not be blocked from sharing content, but the system will help identify and address inappropriate material. Keywords: Picture Sharing, Web Framework, Content-Based Filtering, Online Security . CISDI Journal Reference Format Ehigiator Egho-Promise & Mensah Sitti (2024 Web Framework Content Based Filtering for Picture Sharing. Computing, Information Systems, Development Informatics & Allied Research Journal. Vol 15 No 1, Pp 29--46. dx.doi.org/10.22624/AIMS/CISDI/V15N1P3. Available online at www.isteams.net/cisdijournal

Filtenna Designed with Defected Ground Structure (DGS) for Ultra-wideband Applications

In microwave imaging applications, filter and antenna are the key components as front-end devices and function independently. Antenna radiates and receives signals to or from nearby scattered objects while filter is used to suppress unwanted signals noise before and after the required bandwidth. Current antennas suffer from high loss, bandwidth limitation and impedance mismatch and deteriorate their performance near the band-edges if connected as a stand-alone device. Due to the current trend towards simplicity and size reduction, researchers are focusing on integrating the filter and antenna into a single module called integrated filter-antenna (IFA) or filtering antenna (filtenna). These would improve the noise performance of the system and pre-filtering requirements such as complexity algorithm in inverse scattering techniques This paper introduces a novel contribution in the field of UWB antenna design by incorporating Defected Ground Structure (DGS) on both the antenna and filter. Thus, the main aim of this research is to conduct detail parametric studies of the proposed integrated filter-antenna with defected ground structure (DGS) to enhance the performance of imaging system. The proposed IFA will be analysed on Rogers RO4003C dielectric substrate by using EM tool, Computer Simulation Technology (CST) and measured using R&S Vector Network Analyzer (VNA). A compact ultra-wideband (UWB) filter and UWB elliptical antenna are designed and examined in detail before combining the devices. Different types of DGS are designed and act as the ground layer of the proposed filtering antenna. The bandwidth of each design is then compared with and without the existence of DGS. The results show that the proposed IFA with DGS implementation achieve the targeted objective, with compact size, enhanced bandwidth and better performance compared to the conventional design of filter and antenna. By integrating filter and antenna into one subsystem, it can help to reduce the loss and enhancing the bandwidth of the system thus letting the antenna to operate at more different frequencies that fall within the range. Both simulated and measured results prove that by integrating filter and antenna into one module, a low loss and larger bandwidth can be accomplished. High performance compact IFA can act as microwave transceiver to improve the overall performance of the microwave imaging system, MIS.

FFT-Based Dynamic Token Mixer for Vision

Multi-head-self-attention (MHSA)-equipped models have achieved notable performance in computer vision. Their computational complexity is proportional to quadratic numbers of pixels in input feature maps, resulting in slow processing, especially when dealing with high-resolution images. New types of token-mixer are proposed as an alternative to MHSA to circumvent this problem: an FFT-based token-mixer involves global operations similar to MHSA but with lower computational complexity. However, despite its attractive properties, the FFT-based token-mixer has not been carefully examined in terms of its compatibility with the rapidly evolving MetaFormer architecture. Here, we propose a novel token-mixer called Dynamic Filter and novel image recognition models, DFFormer and CDFFormer, to close the gaps above. The results of image classification and downstream tasks, analysis, and visualization show that our models are helpful. Notably, their throughput and memory efficiency when dealing with high-resolution image recognition is remarkable. Our results indicate that Dynamic Filter is one of the token-mixer options that should be seriously considered. The code is available at https://github.com/okojoalg/dfformer

Multi-modal remote sensing image fusion method guided by local extremum maps-guided image filter

Multi-modal remote sensing image fusion method guided by local extremum maps-guided image filter

Online Defect Detection in LGA Crystallization Imaging Using MANet-Based Deep Learning Image Analysis

In this paper, a MANet-based image detection approach is designed to inspect crystal defects during the cooling crystallization process, like that involving β-form L-glutamic acid (LGA), utilizing an online imaging device. The steps in the presented strategy encompass crystal image preprocessing, crystal image segmentation, and crystal classification. Firstly, the guided image filter is introduced to preprocess the collected crystallization images for offline training and online detection. Then, by using an image augmentation strategy to enlarge the number of crystal image samples for training, the MANet-based network is improved for crystal image segmentation. Accordingly, by defining some features, needle-like crystals can be categorized into four types with an efficient classifier for the detection of normal and defective crystals. The experimental results for the batch crystallization of β-form LGA are provided to illustrate the validity of the presented detection methodology.

Multilevel Ferroelectric Domain Wall Memory for Neuromorphic Computing

AbstractLow‐power and parallel processing Neuromorphic computing that imitates on the human brain's extraordinary data processing and learning capabilities is promising in non‐von Neumann application platforms in the post‐Moore era. Ferroelectric domain walls appearing as nanoscale topological defects in solids exhibit coexisting ordering parameters besides the spontaneous polarization, leading to the emergence of rich physics and electronic effects in the development of neuromorphic electronics that go beyond the conventional CMOS. In this study, the four‐state domain wall memory is developed on LiNbO3 ferroelectric thin films integrated with Si substrates, where each state can be stably manipulated at a specific low voltage, showcasing outstanding fatigue resistance and retention performance. The constructed crossbar array that functions as a kernel for image convolution can be used to implement processing operations like image filtering, sharpness enhancement, and edge detection. Additional simulation from a convolutional neural network model shows 96.98% accuracy on the Modified National Institute of Standards and Technology handwritten digits dataset. The stable multi‐state domain wall memory provides a new approach for computing and promotes research in domain wall electronics to meet future enormous demands of big data.

Enhancing Diagnostic Precision: Evaluation of Preprocessing Filters in Simple Diffusion Kurtosis Imaging for Head and Neck Tumors.

Background: Our initial clinical study using simple diffusion kurtosis imaging (SDI), which simultaneously produces a diffusion kurtosis image (DKI) and an apparent diffusion coefficient map, confirmed the usefulness of SDI for tumor diagnosis. However, the obtained DKI had noticeable variability in the mean kurtosis (MK) values, which is inherent to SDI. We aimed to improve this variability in SDI by preprocessing with three different filters (Gaussian [G], median [M], and nonlocal mean) of the diffusion-weighted images used for SDI. Methods: The usefulness of filter parameters for diagnosis was examined in basic and clinical studies involving 13 patients with head and neck tumors. Results: The filter parameters, which did not change the median MK value, but reduced the variability and significantly homogenized the MK values in tumor and normal tissues in both basic and clinical studies, were identified. In the receiver operating characteristic curve analysis for distinguishing tumors from normal tissues using MK values, the area under curve values significantly improved from 0.627 without filters to 0.641 with G (σ = 0.5) and 0.638 with M (radius = 0.5). Conclusions: Thus, image pretreatment with G and M for SDI was shown to be useful for improving tumor diagnosis in clinical practice.

Transport of the magnetic flux away from a decaying sunspot via convective motions

The interaction between magnetic fields and convection in sunspots during their decay process remains poorly understood,， whereas the formation of sunspots is relatively well studied and fully modeled. Works on the velocity scales at the solar surface have pointed to the existence of the family of granules, whose interaction with the magnetic field leads to the formation of supergranules and their networks, which are visible at the solar surface. The aim of this paper is to consider relationship between the decay of sunspots and convection via the motion of the family of granules and how the diffusion mechanism of magnetic field operates in a decaying sunspot. We report the decay of a sunspot observed by the 1.6m Goode Solar Telescope (GST) with the TiO Broadband Filter Imager (BFI) and the Near-InfraRed Imaging Spectropolarimeter (NIRIS). The analysis was aided by the Helioseismic and Magnetic Imager (HMI) on board the Solar Dynamic Observatory (SDO). In the first step, we followed the decay of the sunspot with HMI data over three days by constructing its evolving area and total magnetic flux. In the second step, the high spatial and temporal resolution of the GST instruments allowed us to analyze the causes of the decay of the sunspot. Afterward, we followed the emergence of granules in the moat region around the sunspot over six hours. The evolution of the trees of fragmenting granules (TFGs) was derived based on their relationship with the horizontal surface flows. We find that the area and total magnetic flux display an exponential decrease over the course of the sunspot decay. We identified 22 moving magnetic features (MMFs) in the moats of pores, which is a signature of sunspot decay through diffusion. We note that the MMFs were constrained to follow the borders of TFGs during their journey away from the sunspot. The TFGs and their development contribute to the diffusion of the magnetic field outside the sunspot. The conclusion of our analysis shows the important role of the TFGs in sunspot decay. Finally, the the family of granules evacuates the magnetic field.

Effective augmentation of front opening unified pod filter images

The front opening unified pods (FOUPs) are popular carriers in semi-conductor industry, and their automatic optical inspection relies on the sufficiency of images. Filter papers at the bottom of FOUP require frequent inspection, while obtaining their images is time-consuming. This work thus investigates augmenting images of filter papers using conventional interpolation and a recent generative adversarial network with variations, including numbers of epochs, usage of progressive learning, loss functions, and cross frame calibration. Drawbacks of interpolated images are exhibited. Results of sixteen models are quantitatively evaluated using the Fréchet inception distance (FID). The best FID among all models is fid = 0.102, and the distribution of generated images is close to that of real images. A user study is also conducted to investigate the authenticity of generated images. Our experiments reveal that given a limited number of training images, after proper geometric calibration, images synthesized by a generative adversarial network are qualified for data augmentation and substantially benefit industrial image classification tasks.

Automatic Speckle Noise Reduction in SAR Imagery Using Optimized Kernel Support Vector Machine

Speckle noise (SN) is one of the major types of noise that frequently occurs in different coherent imaging systems such as medical imaging, Synthetic Aperture Radar (SAR) and active Radar images. SAR is a powerful imaging technology that generates fine-resolution images and monitors the earth’s surface in order to identify its physical properties. The satellite images captured by SAR are mainly affected by SN, which reduces the quality of images and complicates the image representation. Therefore, removing SN from SAR images is one of the major challenges and needs significant attention. The proposed study introduces an optimal Machine Learning (ML) classifier named Kernel Support Vector Machine-Improved Aquila Optimization (KSVM-IAO) for reducing SN in SAR images. This study uses a two-step process called filtering and enhanced despeckling to minimize the consequence of speckle suppression. In the first step, different imaging filters, namely Improved Lee Filter (ILF), Improved Frost Filter (IFF), Improved Kuan Filter (IKF) and Improved Boxcar Filter (IBF), are utilized to remove the SN in SAR images. Next, the denoised image is fed to the second stage, which makes use of an optimized KSVM-IAO classifier to obtain an enhanced despeckle image. The hyperparameters of KSVM are tuned using the IAO algorithm, which reduces overfitting issues and increases accuracy. MATLAB is the simulation tool used for the analysis of SAR images. The simulation outcomes reveal that the proposed KSVM-IAO method obtained excellent SN removal while preserving the edges and fine details with low computational complexity.

Steering Kernel Weighted Guided Image Filtering with Gradient Constraint

In order to make the weighted guided image filter (WGIF) more fully utilize the edge direction of the image, a steering kernel weighted guided image filter (SKWGIF) is proposed by employing the steering kernel to learn the edge direction and incorporating the learning results into the WGIF. However, SKWGIF does not provide a good compromise between the two possibly contradictory objectives of edge-preserving and smoothing, where the image edges are inevitably smoothed. To overcome the drawback, a SKWGIF with gradient constraint (GC-SKWGIF) is proposed by introducing the gradient constraints into the SKWGIF. The gradient constraints allow the filter to take into account the gradient variation of the edges in the filtering process, and therefore the image edges can be better preserved. To verify the effectiveness of the proposed filter, the GC-SKWGIF is applied to edge-aware smoothing, tone mapping of high range dynamic images, image denoising and haze removal. Both theoretical analysis and experimental results show that the proposed filter can produce good resultant images.

Mitigating Viewer Impact From Disturbing Imagery Using AI Filters: A User-Study

Exposure to disturbing imagery can significantly impact individuals, especially professionals who encounter such content as part of their work. This article presents a user study, involving 107 participants, predominantly journalists and human rights investigators, that explores the capability of Artificial Intelligence (AI)-based image filters to potentially mitigate the emotional impact of viewing such disturbing content. We tested five different filter styles, both traditional (Blurring and Partial Blurring) and AI-based (Drawing, Colored Drawing, and Painting), and measured their effectiveness in terms of conveying image information while reducing emotional distress. Our main findings suggest that the AI-based Drawing style filter demonstrates the best performance, offering a promising solution for reducing negative feelings (−30.38%) while preserving the interpretability of the image (97.19%). Overall, this article contributes to the development of a more ethically considerate and effective visual environment for professionals routinely engaging with potentially disturbing imagery.