Subjective Visual Quality Research Articles

The influence of pupil diameter upon and subjective quality of vision following implantable collamer lens (ICL V4c) implantation: An observational study.

This study aimed to assess the change in pupil size and its influence on subjective quality of vision (QoV) in subjects with implanted collamer lenses (ICLs). This retrospective study assessed 53 participants (53 eyes) implanted with ICL (V4c) and categorized them into incremental groups according to pupil diameter. Preoperative and postoperative photopic and mesopic pupil diameter, uncorrected distance visual acuity (UDVA), and QoV questionnaire scores were assessed and compared. Postoperatively, at 3 months, UDVA was -0.10 ± 0.06 logarithm of the minimum angle of resolution (LogMAR), and mean QoV for day and night was 9.34 ± 0.76 and 8.58 ± 1.29, respectively. The mean mesopic and photopic pupil diameters were 6.59 ± 0.79 mm and 4.61 ± 0.74 mm, respectively. Photopic pupil diameter negatively correlated with "QoV day" (Rs = -0.413, P = .001), positively correlated with "haloes" (Rs = 0.568*, P < .001) and "blurred vision" (Rs = 0.243, P = .04) respectively. Mesopic pupil diameter negatively correlated with "QoV night" (Rs = -0.426, P = .001), positively correlated with "haloes" (Rs = 0.624*, P < .001), "starburst" (Rs = 0.233, P = .046) and "difficulty focusing" (Rs = 0.27, P = .025), respectively. Participants had excellent VA at 3-month follow-up. Photopic and mesonic pupil diameter negatively correlated with QoV day and QoV night scores, respectively. Pupil diameter was found to have a more significant effect on visual symptoms at night, and lower QoV due to larger pupil size was more noticeable at night. Further investigation is needed to explore the importance of pupil diameter and its impact on the QoV in ICL implanted patients.

Comparison of cataract patients with regular corneal astigmatism after implantation of extended range-of-vision and bifocal toric intraocular lenses.

To compare the postoperative visual acuity and visual quality between extended range-of-vision and multifocal toric intraocular lens (IOLs) after implantation in cataract patients with regular corneal astigmatism. Department of Ophthalmology, the Second Hospital of Jilin University, Changchun, Jilin Province, China. Retrospective and single-center study. The study involved implanting the Tecnis Symphony (ZXR00IOL) or the bifocal toric (ZMTIOL) in patients undergoing cataract surgery. Three months after surgery, lens performance was evaluated using distance, intermediate, and near visual acuity tests, defocus curves, the modulation transfer function (MTF), a visual function index questionnaire (VF-14), and the adverse optical interference phenomena. The 3-month postoperative follow-up found that both groups had good corrected distance vision. The ZMT group had better-uncorrected distance visual acuity and near visual acuity (p < 0.05). However, the ZXR group showed better uncorrected intermediate visual acuity (p < 0.05) and visual continuity. Overall astigmatism in the postoperative ZMT group was significantly lower than that in the pre-operative group (p < 0.05). The ZMT group had lower total high-order aberrations (tHOs), higher MTF values, and higher VF-14 scores (p < 0.05). Finally, the ZXR group exhibited reduced halo and glare phenomena (p < 0.05). We found that ZMT can effectively correct a corneal astigmatism of 1.0-1.5 D and ZXR can improve patient outcomes regarding subjective optical quality and range of vision. These findings have the potential to improve future astigmatism treatment options.

A Two-Stage Three-Dimensional Attention Network for Lightweight Image Super-Resolution

In recent years, single image super-resolution (SISR) methods using convolutional neural networks (CNN) have achieved satisfactory performance. Nevertheless, the large model scale and the slow inference speed of these methods greatly limit the application scenarios. In this paper, we propose a two-stage three-dimensional attention network (ATTNet) for lightweight image super-resolution. First, we put forward the spatial feature encoder–decoder (SFE-D) with a spatial attention mechanism. Next, the channel transposed attention module (CTAM) with a channel self-attention mechanism is designed. Both the modules are used for fine feature extraction in the low-resolution stage. Finally, the content-based pixel recombination module (CPRM) is proposed to reconstruct the detailed content with a joint attention mechanism in the high-resolution stage. According to our experimental results, significant performance in terms of the quantitative metrics and the subjective visual quality can be achieved on average compared with the state-of-the-art lightweight SISR algorithms.

Результаты имплантации интраокулярных линз с расширенной глубиной фокуса

Purpose. Analysis of clinical and functional results and the subjective assessment by patients of the quality of visual functions after IOL implant with an extended depth of focus (EDOF) of different designs. Material and methods. The study included 29 patients (45 eyes). DFT015 Vivity IOL (Alcon) were implanted in 11 patients, LS-313 MF-15 Lentis Comfort IOL (Teleon) were implanted in 18 patients. Bilateral Vivity implantation in 5 patients, Lentis Comfort – in 11 patients. Monolateral Vivity implantation – 6, Lentis Comfort – 7. Toric models of IOL were implanted in patients with clinically significant astigmatism. Patients were examined before surgery, on the first day after surgery, one month later and three months after surgery. During the examination 3 months after the operation, the visometry of the operated eye was performed for near, at an intermediate distance and in the distance. In patients with bilateral IOL implantation, visometry was performed both monocularly and binocularly. The subjective quality of vision was carried out according to the Visual Function – 14 questionnaire. Results. Phacoemulsification with the implantation of EDOF IOL leads to a significant increase vision acuity in the distance, at an average distance and near, which is especially evident after bilateral implantation and allows patients to perform most of the usual actions without glasses. An increase in visual acuity both at a distance and at an intermediate distance and near was noted by all patients, regardless of the implanted IOL model. The level of satisfaction of patients with their vision after surgery is higher in male patients and with bilateral intervention. The smallest error in calculating the EDOF IOL is marked by the Kane formula. Conclusion. This paper presents a comparison of the clinical and functional results of phacoemulsification with implantation of IOL with EDOF of different designs. The presented data indicate that patients can perform most everyday tasks after these operations without additional correction with glasses. Keywords: cataract, phacoemulsification, IOL with extended depth of focus, EDOF.

Visual results and subjective satisfaction after implantation of two different trifocal diffractive intraocular lenses models (AcrySof IQ PanOptix and AT LISA tri 839 MP).

This study aimed to compare visual results following the implantation of two distinct diffractive trifocal IOLs utilizing various vision evaluation techniques and subjective satisfaction. The study included 32 eyes receiving AcrySof IQ PanOptix implantation and 32 receiving AT LISA tri 839 MP. Post-operative examination at one year included uncorrected visual acuity (UCVA), distance-corrected visual acuity (DCVA) at 4 m, 80 cm, 60 cm, and 40 cm, Contrast sensitivity under photopic and mesopic conditions, and a questionnaire on subjective quality of vision, optical phenomena, and spectacle independence. The two groups were similar in baseline demographics, pre-operative visual, biometric data (except for axial length), and surgical paraments. The PanOptix outperformed the AT LISA tri in the intermediate range when using the standard visual acuity testing (UCVA: 0.23 ± 0.14 logMAR vs. 0.31 ± 0.17 logMAR, p = 0.04; BCVA: 0.21 ± 0.12 vs. 0.28 ± 0.14, p = 0.03) and in the intermediate and far range when using the interactive computer program (intermediate: UNVA "E" optotype test: 0.05 ± 0.12 vs. 0.12 ± 0.13, p = 0.03; Far: UNVA "E" optotype: 0.07 ± 0.12 vs. 0.13 ± 0.10, p = 0.05). In addition, the PanOptix group had higher contrast sensitivity under photopic and mesopic conditions with and without glare at all spatial frequencies. There was no difference in subjective outcomes between the groups. Patients who received PanOptix IOL had better visual outcomes for intermediate and far with higher contrast sensitivity under various conditions. Based on patients ' needs and demands, these results could support discussion when tailoring the optimal multifocal IOl.

Optical Quality and Visual Performance With Different Toric Contact Lens Designs.

To evaluate the optical quality of different toric contact lens (CL) designs and compare their on-eye visual correction performance. Twenty soft CL wearers aged 20 to 39 years were enrolled. Two daily disposable silicone-hydrogel toric CLs were tested: the "Eyelid Stabilized Design" (ESD-CL) and prism-ballast design (PB-CL); a spherical daily disposable silicone hydrogel CL (spherical CL) was used as a control. On-eye performance was compared for corrected distance visual acuity (CDVA), astigmatism, and ocular higher-order aberrations (HOAs); astigmatism and ocular HOAs were measured with a wavefront sensor. The subjective quality of vision, rated for "blurred vision" and "double vision," lens rotation, and fitting were also compared. The ESD-CLs, PB-CL, and no-CL provided better CDVA than spherical CL ( P <0.05). Compared with spherical CL and no CL, PB-CL and ESD-CLs caused significantly lesser astigmatism ( P <0.05). Coma was significantly lesser with ESD-CLs than that with PB-CL ( P <0.05); total HOAs did not differ among the four conditions. The subjective ratings for blurred and double vision were significantly lower with ESD-CLs than those with spherical CL ( P <0.05). Toric CLs provide a better CDVA than spherical CLs. However, differences in coma and subjective symptoms may occur because of the design of toric CLs.

Visual performance and subjective experience 3 and 12 months after combined implantation of 2 new complementary continuous phase multifocal IOLs.

To assess the 3-month and 12-month postoperative visual performance and subjective quality of vision (QoV) after combined implantation of complementary continuous phase multifocal IOLs. Private Practice, United Kingdom. Case series. The study enrolled 44 patients undergoing phacoemulsification with implantation of an Artis Symbiose Mid (Cristalens, France) in the dominant eye and an Artis Symbiose Plus (Cristalens, France) in the nondominant eye. Refraction, uncorrected (UDVA) and best-corrected (CDVA) distance visual acuities, uncorrected intermediate (UIVA) and near (UNVA) visual acuities, electronic reading desk, and a QoV questionnaire were evaluated 3 and 12 months postoperatively. The mean binocular UDVA was -0.06 ± 0.08 logMAR and -0.07 ± 0.06 logMAR at 3 and 12 months (P=0.097). The mean binocular UIVA was 0.03 ± 0.13 logMAR and 0.03 ± 0.10 logMAR (P =1.0) respectively. Mean binocular UNVA was 0.07 ± 0.10 logMAR and 0.07 ± 0.08 logMAR, (P =0.875) respectively. There was a significant improvement in QoV for both day and night between 3 and 12 months, with a significant reduction in haloes at 12 months. Spectacle independence was reported in 93.2% of cases at 12 months. The Artis Symbiose Mid and Plus IOLs combined implantation provided an excellent range of unaided vision at 3 and 12 months. There was a significant improvement in QoV and less haloes at 12 months. This IOL combination provided very high rates of complete spectacle independence.

Comparative analysis of visual outcomes with 2 enhanced monofocal intraocular lenses.

To compare the visual performance, spectacle independence, and subjective visual quality of 2 enhanced monofocal intraocular lenses (IOLs): Eyhance ICB00 and ZOE Primus-HD. Ophthalmology Department, San Raffaele Scientific Institute, Milan, Italy. Prospective case series. Patients without ocular comorbidities and corneal astigmatism <0.75 diopters (D) undergoing cataract surgery with bilateral implantation of enhanced monofocal Eyhance and ZOE were included. 6 months postoperatively, the following parameters were analyzed: subjective and objective refraction; monocular and binocular corrected (4 m) and uncorrected (UDVA) distance visual acuities; corrected distance, intermediate (66 cm), and near (40 cm) visual acuities, as well as uncorrected intermediate (UIVA) and uncorrected near (UNVA) visual acuities; photopic contrast sensitivity; binocular defocus curve; halo and glare perception; and spectacle independence. 100 eyes of 50 patients were evaluated (25 patients per IOL group). The 2 IOLs showed superimposable visual performances, without significant differences in terms of refractive and visual outcomes, defocus curves, contrast sensitivity, scores of vision quality, and spectacle independence. Of note, monocular and binocular UDVA was excellent in both groups. The 2 IOL models achieved satisfactory binocular UIVA, with more than 70% of patients reaching a binocular UIVA of ≤0.1 logMAR. Up to 84% of patients eventually declared to feel comfortable frequently while staying at intermediate distance. The general visual outcome of the Eyhance ICB00 and the ZOE Primus-HD enhanced monofocal IOLs is similar, with special reference to the satisfactory spectacle independence for intermediate distance.

Non-Uniform-Illumination Image Enhancement Algorithm Based on Retinex Theory

To address the issues of fuzzy scene details, reduced definition, and poor visibility in images captured under non-uniform lighting conditions, this paper presents an algorithm for effectively enhancing such images. Firstly, an adaptive color balance method is employed to address the color differences in low-light images, ensuring a more uniform color distribution and yielding a low-light image with improved color consistency. Subsequently, the image obtained is transformed from the RGB space to the HSV space, wherein the multi-scale Gaussian function is utilized in conjunction with the Retinex theory to accurately extract the lighting components and reflection components. To further enhance the image quality, the lighting components are categorized into high-light areas and low-light areas based on their pixel mean values. The low-light areas undergo improvement through an enhanced adaptive gamma correction algorithm, while the high-light areas are enhanced using the Weber–Fechner law for optimal results. Then, each block area of the image is weighted and fused, leading to its conversion back to the RGB space. And a multi-scale detail enhancement algorithm is utilized to further enhance image details. Through comprehensive experiments comparing various methods based on subjective visual perception and objective quality metrics, the algorithm proposed in this paper convincingly demonstrates its ability to effectively enhance the brightness of non-uniformly illuminated areas. Moreover, the algorithm successfully retains details in high-light regions while minimizing the impact of non-uniform illumination on the overall image quality.

Effects of Different Full-Reference Quality Assessment Metrics in End-to-End Deep Video Coding

Visual quality assessment is often used as a key performance indicator (KPI) to evaluate the performance of electronic devices. There exists a significant association between visual quality assessment and electronic devices. In this paper, we bring attention to alternative choices of perceptual loss function for end-to-end deep video coding (E2E-DVC), which can be used to reduce the amount of data generated by electronic sensors and other sources. Thus, we analyze the effects of different full-reference quality assessment (FR-QA) metrics on E2E-DVC. First, we select five optimization-suitable FR-QA metrics as perceptual objectives, which are differentiable and thus support back propagation, and use them to optimize an E2E-DVC model. Second, we analyze the rate–distortion (R-D) behaviors of an E2E-DVC model under different loss function optimizations. Third, we carry out subjective human perceptual tests on the reconstructed videos to show the performance of different FR-QA optimizations on subjective visual quality. This study reveals the effects of the competing FR-QA metrics on E2E-DVC and provides a guide for further future study on E2E-DVC in terms of perceptual loss function design.

Effect of postoperative residual astigmatism on visual outcomes after trifocal intraocular lens implantation.

The aim of this study was to evaluate the effect of residual astigmatism on postoperative visual outcomes after trifocal intraocular lens implantation. In this prospective observational study, we divided 156 eyes into two groups according to postoperative astigmatism measured by subjective optometry and followed them up for 3 months. Visual acuity, modulation transfer function (MTF) curves, Strehl ratio (SR), Visual Function Index-14 scores, and photic phenomena were compared. Linear regression analysis revealed a weak correlation between residual astigmatism and uncorrected distance visual acuity (UDVA) (r = 0.190, P = 0.016) at 3 months and a significant between-group difference at 1- and 3-month postoperative UDVA (P = 0.038, P = 0.018, respectively). MTF curve values and SR (MTF-10 total, MTF-10 cornea, MTF-30 total, MTF-30 cornea, SR Total, and SR cornea) were significantly worse (P < 0.001), and the Visual Function Index-14 scores were lower in the 0.5 < astigmatism ≤ 1.25 D group (P < 0.05) than in the astigmatism ≤ 0.5 D group. No significant differences were found in the frequency, severity, and bothersomeness of photic phenomena (P > 0.05). Postoperative residual astigmatism affects the UDVA of the trifocal intraocular lens-implanted eyes. Although we found no significant differences in uncorrected intermediate and near visual acuity, both objective and subjective visual quality were affected, suggesting the need for surgical planning when the anticipated postoperative astigmatism is >0.5 D.

A comparison of the visual performance of refractive rotationally asymmetric multifocal intraocular lenses and trifocal intraocular lenses

Objective: The aim of this study is to compare the short term visual outcomes and subjective visual quality of the monocular implantation of a refractive rotationally asymmetric multifocal intraocular lens (IOL) (SBL-3, Lenstec, Inc., USA) with those of a trifocal IOL (AT LISA tri 839 MP, Carl Zeiss Meditec, Jena, Germany). Methods: This study involved a total of 46 patients (46 eyes) with age-related cataracts, who underwent uncomplicated cataract surgery. Twenty-two patients (22 eyes) had SBL-3 multifocal IOLs implanted (group A) and 24 patients (24 eyes) had AT LISA tri 839 MPs implanted (group B). The patients were all evaluated three months after surgery. Results: Group B had a significantly better performance (0.127±0.070) logMAR (20/40 Snellen equivalent) than group A (0.25±0.093) LogMAR for monocular UIVA at 80 cm (p &lt; 0.05). There was no statistically significant difference at other distances. The monocular defocus curves showed that group B had a better performance than group A at a vergence of −1.5 D (p&lt;0.05). Higher-order aberrations (HOAs) were significantly higher in group A (p &lt;0.05). The SBL-3 IOLs resulted in much lower MTF cut-off (23.68±9.892) cpd and Strehl ratio (0.13±0.052) than trifocal IOLs (MTF cut-off 26.25±8.751 cpd, Strehl ratio 0.15±0.063). But there was no significant difference between the two IOLs (p &gt;0.05). Complaints of halos at night were made by 37.5% (9/24) of group A and 9.1% (2/22) of group B, but they did not interfere with their vision. All the patients reported a high level of satisfaction and said they no longer needed to use spectacle correction at either far, intermediate or near distances. Conclusions: Both SBL-3 IOLs and AT LISA tri 839 MP trifocal IOLs provided good short-term visual outcomes at all distances postoperatively with a high level of spectacle independence and patient satisfaction. The AT LISA tri 839 MP lens provided better intermediate vision at 80 cm, while similar visual acuity (VA) was reported at other distances.

Application of Principal Component Analysis to advancing digital phenotyping of plant disease in the context of limited memory for training data storage

Despite its widespread employment as a highly efficient dimensionality reduction technique, limited research has been carried out on the advantage of Principal Component Analysis (PCA)–based compression/reconstruction of image data to machine learning-based image classification performance and storage space optimization. To address this limitation, we designed a study in which we compared the performances of two Convolutional Neural Network-Random Forest Algorithm (CNN-RF) guava leaf image classification models developed using training data from a number of original guava leaf images contained in a predefined amount of storage space (on the one hand), and a number of PCA compressed/reconstructed guava leaf images contained in the same amount of storage space (on the other hand), on the basis of four criteria – Accuracy, F1-Score, Phi Coefficient and the Fowlkes–Mallows index. Our approach achieved a 1:100 image compression ratio (99.00% image compression) which was comparatively much better than previous results achieved using other algorithms like arithmetic coding (1:1.50), wavelet transform (90.00% image compression), and a combination of three transform-based techniques – Discrete Fourier (DFT), Discrete Wavelet (DWT) and Discrete Cosine (DCT) (1:22.50). From a subjective visual quality perspective, the PCA compressed/reconstructed guava leaf images presented almost no loss of image detail. Finally, the CNN-RF model developed using PCA compressed/reconstructed guava leaf images outperformed the CNN-RF model developed using original guava leaf images by 0.10% accuracy increase, 0.10 F1-Score increase, 0.18 Phi Coefficient increase and 0.09 Fowlkes–Mallows increase.

Higher-order aberrations and visual quality after incision lenticule extraction surgery with intraoperative angle kappa adjustments between small and large kappa patients: A 2-year follow-up.

To evaluate the postoperative visual outcomes, that is, corneal higher-order aberrations (HOAs) and visual quality, of patients with an angle kappa greater than 0.30 mm who underwent angle kappa adjustment during small-incision lenticule extraction (SMILE) 2 years after surgery compared to eyes with an angle kappa less than 0.30 mm. This was a retrospective study and included 12 patients from October 2019 to December 2019 who underwent the SMILE procedure for correction of myopia and myopic astigmatism and had one eye with a large kappa angle and another eye with a small kappa angle. Twenty-four months after surgery, an optical quality analysis system (OQAS II; Visiometrics, Terrassa, Spain) was used to measure the modulation transfer function cutoff frequency (MTFcutoff), Strehl2D ratio, and objective scatter index (OSI). HOAs were measured with a Tracey iTrace Visual Function Analyzer (Tracey version 6.1.0; Tracey Technologies, Houston, TX, USA). Assessment of subjective visual quality was achieved using the quality of vision (QOV) questionnaire. At 24 months postoperatively, the mean spherical equivalent (SE) refraction was - 0.32 ± 0.40 and - 0.31 ± 0.35 in the S-kappa group (kappa <0.3 mm) and the L-kappa group (kappa ≥0.3 mm), respectively (P > 0.05). The mean OSI was 0.73 ± 0.32 and 0.81 ± 0.47, respectively (P > 0.05). There was no significant difference in MTFcutoff and Strehl2D ratio between the two groups (P > 0.05). Total HOA, coma, spherical, trefoil, and secondary astigmatism were not significantly different (P > 0.05) between the two groups. Adjustment of angle kappa during SMILE helps reduce the decentration, results in less HOAs, and promotes visual quality. It provides a reliable method to optimize the treatment concentration in SMILE.

A 3-month comparison study of subjective and objective visual quality of small incision lenticule extraction and transepithelial photorefractive keratectomy in patients with low and moderate myopia.

To compare the subjective and objective visual quality between small incision lenticule extraction (SMILE) and transepithelial photorefractive keratectomy (tPRK) in patients with low and moderate myopia. Patients undertaking SMILE or tPRK for the correction of low and moderate myopia were consecutively recruited in this prospective cohort study with a 3-month follow-up period. Objective evaluation [visual acuity test, manifest refraction, wavefront aberrations, the total cut-off value of the total modulation transfer function (MTFcut-off), and Strehl ratio (SR)] and subjective evaluation of visual quality (quality-of-life questionnaire) were conducted before surgery and at days 1, 7, 30, and 90 after surgery. A total of 47 patients (94 eyes) with SMILE and 22 patients (22 eyes) with tPRK were enrolled. The uncorrected visual acuity (UCVA) was better in SMILE patients on day 7 after surgery (1.13±0.13 vs 0.99±0.17, t=4.85, P<0.001) but was comparable at days 30 and 90. At day 90, the SMILE group had a lower spherical equivalent (SE) than the tPRK group (0.04±0.31 vs 0.19±0.43, t=2.08, P=0.042). Total higher order aberrations (HOAs) were induced in both surgical types, which were more evident in the tPRK group with 3-mm pupil diameter (0.16±0.07 vs 0.11±0.05, t=4.27, P<0.001) and 5-mm pupil diameter (0.39±0.17 vs 0.36±0.11, t=2.33, P=0.022). The MTFcut-off and SR showed a trend of improvement in both SMILE and tPRK patients but were statistically better in the SMILE group with both pupil diameters. There was a significant improvement of contrast sensitivity (CS) over baseline levels at the spatial frequency of 18 cycles/degree (c/d) in the SMILE group (F=2.72, P=0.033) and at 3 c/d (F=3.03, P=0.031), 12 c/d (F=3.72, P=0.013), and 18 c/d (F=4.62, P=0.004) in the tPRK group. The subjective quality of life questionnaire showed a steady improvement in the SMILE group (F=8.31, P<0.001) but not the tPRK group. SMILE and tPRK are both safe and effective ways to correct low and moderate myopia. A generally better and quicker recovery of visual quality favors the application of SMILE in qualified patients.

Reparameterizable Multibranch Bottleneck Network for Lightweight Image Super-Resolution.

Deployment of deep convolutional neural networks (CNNs) in single image super-resolution (SISR) for edge computing devices is mainly hampered by the huge computational cost. In this work, we propose a lightweight image super-resolution (SR) network based on a reparameterizable multibranch bottleneck module (RMBM). In the training phase, RMBM efficiently extracts high-frequency information by utilizing multibranch structures, including bottleneck residual block (BRB), inverted bottleneck residual block (IBRB), and expand-squeeze convolution block (ESB). In the inference phase, the multibranch structures can be combined into a single 3 × 3 convolution to reduce the number of parameters without incurring any additional computational cost. Furthermore, a novel peak-structure-edge (PSE) loss is proposed to resolve the problem of oversmoothed reconstructed images while significantly improving image structure similarity. Finally, we optimize and deploy the algorithm on the edge devices equipped with the rockchip neural processor unit (RKNPU) to achieve real-time SR reconstruction. Extensive experiments on natural image datasets and remote sensing image datasets show that our network outperforms advanced lightweight SR networks regarding objective evaluation metrics and subjective vision quality. The reconstruction results demonstrate that the proposed network can achieve higher SR performance with a 98.1 K model size, which can be effectively deployed to edge computing devices.

Shearlet-based intensity classification loop filter for video coding

The new Versatile Video Coding standard (VVC) exhibits a significantly higher coding efficiency in comparison to its predecessor High Efficiency Video Coding (HEVC). Since the finalization of VVC in July 2020 many new in-loop compression tools were suggested, especially with data-driven approaches. The question arises if non-data-driven in-loop filtering approaches can further increase coding efficiency beyond the capability of VVC. In this paper we approach the task of filtering images corrupted by quantization noise from the perspective of applied harmonic analysis, an area of applied mathematics focusing on the efficient representation, analysis and encoding of data. The shearlet-based intensity classification loop filter (SCLF) is a novel non-data-driven approach using an overcomplete and sparsifying transform, the shearlet transform. The basic idea is to achieve a signal-noise separation by applying a shearlet transform. Shearlets can identify important structures of natural images such as edges in the sparse transform domain. Each shearlet transform coefficient is classified into different classes. All coefficients with similar intensities are grouped into the same class, which eventually gives a partition of the set of all shearlet coefficients. This separates important information from noise and ensures an accurate reconstruction of original information by performing an inverse shearlet transform combined with Wiener filtering separately for each class. SCLF effectively removes compression artifacts and therefore restores subjective visual quality. Simulation results show that by adding SCLF to the in-loop stage yet further average bit rate reductions of up to 1.5% are achieved for JVET common test sequences.

Comparison of objective and subjective visual quality after flapless laser vision correction for mild to moderate myopia: SMILE vs PRK.

To investigate the differences in surgical results and the objective and subjective quality of vision (QoV) of patients after small incision lenticule extraction (SMILE) versus alcohol-assisted photorefractive keratectomy (PRK). Medical records of patients treated with SMILE and PRK were retrospectively examined. Visual quality, biometric parameters, Strehl ratio (SR), and corneal higher-order aberrations (HOAs) within a 6.0mm area were recorded. The effective optical zone (EOZ) and decentration were measured using a tangential pre-post operation difference map. Subjective QoV and operation satisfaction were evaluated 6months postoperatively using the Quality of Vision questionnaire. The study comprised 100 eyes treated with SMILE (preoperative mean spherical equivalent (SE), - 4.52 ± 0.81 dioptres (D)) and 69 eyes with PRK (mean SE, - 4.21 ± 1.25 D). Six months postoperatively, the EOZ reduction was significantly larger in the PRK group (P < 0.001). Decentrations were comparable between the groups. Regarding visual symptoms, monocular diplopia was more common following PRK (P = 0.02), and 98 (98.00%) SMILE-treated and 67 (97.10%) PRK-treated patients were satisfied with the QoV. Both groups demonstrated significant increases in total HOAs, coma, and spherical aberration (SA) at 6months postoperatively compared to preoperatively (P < 0.001); these values were significantly higher in the PRK (P < 0.05) compared to the SMILE group. SR increased significantly only in the PRK group (P < 0.05). Although EOZ was more consistent with anticipated treatment and HOAs were fewer in SMILE, high patient-reported satisfaction and good corneal optical quality were achieved in both groups, indicating that both SMILE and alcohol-assisted PRK are excellent options for mild to moderate myopia correction.

GPSR: Gradient-Prior-Based Network for Image Super-Resolution

Recent deep learning has shown great potential in super-resolution (SR) tasks. However, most deep learning-based SR networks are optimized via pixel-level loss (i.e., L1, L2, and MSE), which forces the networks to output the average of all possible predictions, leading to blurred details. Especially in SR tasks with large scaling factors (i.e., ×4, ×8), the limitation is further aggravated. To alleviate this limitation, we propose a Gradient-Prior-based Super-Resolution network (GPSR). Specifically, a detail-preserving Gradient Guidance Strategy is proposed to fully exploit the gradient prior to guide the SR process from two aspects. On the one hand, an additional gradient branch is introduced into GPSR to provide the critical structural information. On the other hand, a compact gradient-guided loss is proposed to strengthen the constraints on the spatial structure and to prevent the blind restoration of high-frequency details. Moreover, two residual spatial attention adaptive aggregation modules are proposed and incorporated into the SR branch and the gradient branch, respectively, to fully exploit the crucial intermediate features to enhance the feature representation ability. Comprehensive experimental results demonstrate that the proposed GPSR outperforms state-of-the-art methods regarding both subjective visual quality and objective quantitative metrics in SR tasks with large scaling factors (i.e., ×4 and ×8).

RESEARCH PROGRESS OF DEEP LEARNING IN LOW-DOSE CT IMAGE DENOISING.

Low-dose computed tomography (CT) will increase noise and artefacts while reducing the radiation dose, which will adversely affect the diagnosis of radiologists. Low-dose CT image denoising is a challenging task. There are essential differences between the traditional methods and the deep learning-based methods. This paper discusses the denoising approaches of low-dose CT image via deep learning. Deep learning-based methods have achieved relatively ideal denoising effects in both subjective visual quality and quantitative objective metrics. This paper focuses on three state-of-the-art deep learning-based image denoising methods, in addition, four traditional methods are used as the control group to compare the denoising effect. Comprehensive experiments show that the deep learning-based methods are superior to the traditional methods in low-dose CT images denoising.