Eye Pairs Research Articles

Evolution: Decoding the adaptation of multi-eyed visual systems

Many invertebrates possess more than two pairs of eyes- but does eye redundancy aid in ecological diversification? A new study finds varied size adaptation of different eye pairs in spiders, demonstrating how developmental modularity of multi-eyed systems effectively balances selective pressures.

Allometry and ecology shape eye size evolution in spiders

Eye size affects many aspects of visual function, but eyes are costly to grow and maintain. The allometry of eyes can provide insight into this trade-off, but this has mainly been explored in species that have two eyes of equal size. By contrast, animals possessing larger visual systems can exhibit variable eye sizes within individuals. Spiders have up to four pairs of eyes whose sizes vary dramatically, but their ontogenetic, static, and evolutionary allometry has not yet been studied in a comparative context. We report variable dynamics in eye size across 1,098 individuals in 39 species and 8 families, indicating selective pressures and constraints driving the evolution of different eye pairs and lineages. Supplementing our sampling with a recently published phylogenetically comprehensive dataset, we confirmed these findings across more than 400 species; found that ecological factors such as visual hunting, web building, and circadian activity correlate with eye diameter; and identified significant allometric shifts across spider phylogeny using an unbiased approach, many of which coincide with visual hunting strategies. The modular nature of the spider visual system provides additional degrees of freedom and is apparent in the strong correlations between maximum/minimum investment and interocular variance and three key ecological factors. Our analyses suggest an antagonistic relationship between the anterior and posterior eye pairs. These findings shed light on the relationship between spider visual systems and their diverse ecologies and how spiders exploit their modular visual systems to balance selective pressures and optical and energetic constraints.

Enhanced choroidal thickness, total choroidal area and choroidal vascularity index from deep‐learning generated OCT images

Aims/Purpose: Our goal is to enhance the visualization of the choroid in clinical standard spectral‐domain OCT (SDOCT) scans using a deep‐learning (DL) approach that uses matching swept‐source OCT (SSOCT). We aim to improve the measurement of the sub‐foveal choroidal thickness (SFCT), total choroidal area (TCA), and choroidal vascularity index (CVI) in SDOCT images.Methods: A total of 105 350 SDOCT and SSOCT B‐scan images (Zeiss Cirrus 5000 and Zeiss PlexElite) were used to develop the DL model. The data set included B‐scans from healthy (238 eye pairs from 155 patients) and glaucoma (192 eye pairs from 122 patients) subjects. The DL model learned deep anatomical features from SSOCT scans and generated enhanced SDOCT images. The peak signal‐to‐noise ratio (PSNR) and structural similarity index metric (SSIM) were used to assess the quality of generated images. In addition, we compared the anatomical parameters derived from the generated and ground‐truth SSOCT images, including SVCT, TCA and CVI values.Results: We tested our method using 20 580 B‐scan image pairs from 84 eyes. The DL‐generated images effectively preserved the anatomical structure of the retina layers and improved the visualization of the choroid. The enhanced SDOCT images had higher image quality than the original SDOCT images, as demonstrated by the higher PSNR (31 dB vs. 18 dB) and SSIM values (3.4 times higher SSIM than the original SDOCT images). SVCT, TCA, and CVI measurements from the DL‐generated SDOCT images showed strong correlations with corresponding measurements using SSOCT with high Pearson's coefficients (0.93, 0.99, and 0.93), high intra‐class correlation indices (0.96, 0.99 and 0.96), and high coefficient of determination r2 values (0.85, 0.98 and 0.86).Conclusions: Our DL model enhances the visibility of the choroid in images, surpassing the capabilities of standard clinical SDOCT images. As a result, accurate measurement of SVCT, TCA and CVI, which were previously limited due to imaging depth constraints of SDOCT, becomes possible. This technology opens up new possibilities for utilizing affordable SDOCT devices to study the choroid in healthy and pathological conditions.

Automated Driver Drowsiness Detection

This project uses image processing and Python programming language to create an automatic sleepiness detection system for drivers. The system's objective is to develop an algorithm that can reliably gauge a driver's sleepiness under various circumstances by using real-time grayscale photographs. The driver will be observed by the system as it scans their face and eyelids for hypo vigilance, which is a precursor to tiredness. To properly determine the driver's level of sleepiness, the system entails three steps: feature selection, eye-pair state detection, and decision-making. The algorithm will differentiate between the two states on a collection of captured photos of both tired and awake drivers. To identify sleepiness in drivers, facial feature recognition algorithms has been employed, with an emphasis on the eye pair. Images will be taken with a mounted camera. For the system to successfully be integrated into current driver assistance systems, it will be measured on how well it is able to identify tiredness. The project's importance stem from its ability is to increase road safety by identifying driver sleepiness, a major contributing factor to accidents on the road. In conclusion, this research will significantly impact road safety by tackling the problem of accidents caused by drowsiness.

Identifying the Key Components in ResNet-50 for Diabetic Retinopathy Grading from Fundus Images: A Systematic Investigation.

Although deep learning-based diabetic retinopathy (DR) classification methods typically benefit from well-designed architectures of convolutional neural networks, the training setting also has a non-negligible impact on prediction performance. The training setting includes various interdependent components, such as an objective function, a data sampling strategy, and a data augmentation approach. To identify the key components in a standard deep learning framework (ResNet-50) for DR grading, we systematically analyze the impact of several major components. Extensive experiments are conducted on a publicly available dataset EyePACS. We demonstrate that (1) the DR grading framework is sensitive to input resolution, objective function, and composition of data augmentation; (2) using mean square error as the loss function can effectively improve the performance with respect to a task-specific evaluation metric, namely the quadratically weighted Kappa; (3) utilizing eye pairs boosts the performance of DR grading and; (4) using data resampling to address the problem of imbalanced data distribution in EyePACS hurts the performance. Based on these observations and an optimal combination of the investigated components, our framework, without any specialized network design, achieves a state-of-the-art result (0.8631 for Kappa) on the EyePACS test set (a total of 42,670 fundus images) with only image-level labels. We also examine the proposed training practices on other fundus datasets and other network architectures to evaluate their generalizability. Our codes and pre-trained model are available online.

Choroidal thickness measurements of subjects with pseudoexfoliative syndrome and pseudoexfoliative glaucoma: A contralateral eye study.

In this study, we aimed to investigate the peripapillary choroidal thickness (PPCT) and macular choroidal thickness (MCT) in pseudoexfoliation (PEX) syndrome and compare the data of the contralateral eyes according to the presence of biomicroscopic eye involvement and glaucoma. In this cross-sectional case-control study, PPCT and MCT measurements were analyzed in 162 eyes of 81 subjects with PEX syndrome, diagnosed biomicroscopically with the detection of PEX material. The sample included 63 eyes with pseudoexfoliation glaucoma (PG), 49 eyes with visible PEX material alone without glaucoma (PM), 50 fellow eyes without biomicroscopically visible PEX material (F), and 48 eyes of 24 healthy individuals (controls) without any sign of PEX in the detailed ophthalmologic examination. The PEX syndrome group consisted of 25 PM-F, 25 PG-F, 19 PG-PG, and 12 PM-PM eye pairs. The PPCT and MCT values were compared between the eye-pairs of the subjects with the PEX syndrome. The mean PPCT measurements were 183.3 ± 8.1 µm, 158.5 ± 5.4 µm, 167.8 ± 5.9 µm, and 149.9 ± 5.5 µm for the eyes in the control, F, PM, and PG groups, respectively. The eyes in the PG group had statistically significantly lower measurements than those in the control group (p < 0.01). In the contralateral eye comparison of the subjects with PEX syndrome, no significant difference was found in relation to the mean PPCT and MCT measurements between the PM-F, PG-F, and PG-PG eye pairs (p > 0.05 for all). Although the eye groups with PEX syndrome had lower mean PPCT measurements than the controls, the contralateral eye analysis of the asymmetrically involved eye pairs showed no significant differences.

Discriminative kernel convolution network for multi-label ophthalmic disease detection on imbalanced fundus image dataset

It is feasible to recognize the presence and seriousness of eye disease by investigating the progressions in retinal biological structures. Fundus examination is a diagnostic procedure to examine the biological structure and anomalies present in the eye. Ophthalmic diseases like glaucoma, diabetic retinopathy, and cataracts are the main cause of visual impairment worldwide. Ocular Disease Intelligent Recognition (ODIR-5K) is a benchmark structured fundus image dataset utilized by researchers for multi-label multi-disease classification of fundus images. This work presents a Discriminative Kernel Convolution Network (DKCNet), which explores discriminative region-wise features without adding extra computational cost. DKCNet is composed of an attention block followed by a Squeeze-and-Excitation (SE) block. The attention block takes features from the backbone network and generates discriminative feature attention maps. The SE block takes the discriminative feature maps and improves channel interdependencies. Better performance of DKCNet is observed with InceptionResnet backbone network for multi-label classification of ODIR-5K fundus images with 96.08 AUC, 94.28 F1-score, and 0.81 kappa score. The proposed method splits the common target label for an eye pair based on the diagnostic keyword. Based on these labels, over-sampling and/or under-sampling are done to resolve the class imbalance. To check the bias of the proposed model towards training data, the model trained on the ODIR dataset is tested on three publicly available benchmark datasets. It is observed that the proposed DKCNet gives good performance on completely unseen fundus images also.

Assessment of Efficacy of a Novel Crosslinking Protocol with Intracameral Oxygen (Bubble-CXL) in Increasing the Corneal Stiffness Using Atomic Force Microscopy.

The environmental oxygen level plays a critical role in corneal crosslinking (CXL), a treatment method to increase corneal biomechanical stability. In this study, we introduce a new CXL method (Bubble-CXL), in which intracameral oxygen serves as an additional oxygen source during eye treatment. The efficiency of this new method was compared with the efficiency of the standard CXL method. Three fresh porcine eye pairs were included in this study. One eye of each pair was treated with standard CXL, whereas in the partner eye, intracameral oxygen was injected prior to CXL and removed at the end of the procedure. The Young’s modulus of each cornea was measured using atomic force microscopy. All analyzed corneas treated with intracameral oxygen showed significantly higher Young’s modulus and thus an increased stiffness compared to the cornea of the partner eye treated with the standard protocol. Using intracameral oxygen in CXL therapy may increase crosslinking efficiency and improve biomechanical corneal properties.

Gaze tracker parameters have little association with visual field metrics of intrasession frontloaded SITA-Faster 24-2 visual field results.

PurposeTo determine the usefulness of Humphrey Field Analyser (HFA) SITA‐Faster 24–2 gaze tracker outputs on interpreting intra‐visit visual field (VF) result pairs.MethodsAnalysis of 1380 right–left eye pairs and 1432 pairs of test 1‐test 2 intrasession VF results of patients seen within a university‐based glaucoma service was undertaken to understand gaze deviation distributions. Output gaze tracker results were aggregated into total ticks, sum of amplitudes and average amplitudes. Correlations between visual field indices (mean deviation [MD], “events” and overall hill of vision) and independent variables (age and test order) were performed using one eye from each subject.ResultsThere was no association of test order (right–left, test 1‐test 2) with eye movements. There was a significant, but weak correlation between eye movements and age (r = 0.16). Correlations of eye movements with MD were driven by more severe MD values. There were no significant correlations between intrasession difference in eye movements and the change in MD, number of “events” and hill of vision, or in the root mean square of sensitivity and total deviation values. There was also no significant correlation between gaze tracker outputs and another commonly used “reliability” metric, false positive rate.ConclusionsEye movement parameters as currently reported by the HFA do not appear to be correlated with key sensitivity parameters when considering the repeatability of intrasession SITA‐Faster 24–2 VF results. Thus, current gaze tracker outputs do not appear to provide clinically meaningful information for interpretation of intra‐visit visual field results that cannot already be garnered using other strategies.

Force of lifelong crystalline lens growth: chronic traumatic mechanical insult to the choroid.

To calculate the forces applied to the uvea and retina as a result of lifelong crystalline lens growth. Retrospective study. MRI Research, Inc., Middleburg Heights, Ohio; Institute of Ophthalmology and Visual Science UMDNJ-New Jersey Medical School, Newark, New Jersey; USC Psychology University of Southern California, Los Angeles. Magnetic resonance images were acquired from 15 phakic/pseudophakic eye pairs in patients with cataract (ages 46 to 83 years). Choroidal lengths were measured. The forces required to produce differences between phakic/pseudophakic choroidal lengths were calculated. The length of the choroid is greater in the phakic eye compared with the corresponding pseudophakic eye (n = 15), and this difference increases with age (P = .00006; power = 0.99). The corresponding choroidal strain also increases with age (P = .00003, power = 0.99) as do the forces required to produce such a change in choroidal length (P = .000008, power = 0.99). The authors theorize that lifelong crystalline lens growth applies a chronic, traumatic, mechanical insult to the uvea and retina. This previously unknown, ever-increasing, force appears to stretch the choroidal tissue and may be an intraocular pressure-independent modifiable risk factor for retinal disease. Implications exist for understanding the pathophysiology of retinal diseases in the aging eye that are often comorbid with cataracts, for example, glaucoma, macular degeneration, and diabetic retinopathy.

The visual pathway in sea spiders (Pycnogonida) displays a simple serial layout with similarities to the median eye pathway in horseshoe crabs

BackgroundPhylogenomic studies over the past two decades have consolidated the major branches of the arthropod tree of life. However, especially within the Chelicerata (spiders, scorpions, and kin), interrelationships of the constituent taxa remain controversial. While sea spiders (Pycnogonida) are firmly established as sister group of all other extant representatives (Euchelicerata), euchelicerate phylogeny itself is still contested. One key issue concerns the marine horseshoe crabs (Xiphosura), which recent studies recover either as sister group of terrestrial Arachnida or nested within the latter, with significant impact on postulated terrestrialization scenarios and long-standing paradigms of ancestral chelicerate traits. In potential support of a nested placement, previous neuroanatomical studies highlighted similarities in the visual pathway of xiphosurans and some arachnopulmonates (scorpions, whip scorpions, whip spiders). However, contradictory descriptions of the pycnogonid visual system hamper outgroup comparison and thus character polarization.ResultsTo advance the understanding of the pycnogonid brain and its sense organs with the aim of elucidating chelicerate visual system evolution, a wide range of families were studied using a combination of micro-computed X-ray tomography, histology, dye tracing, and immunolabeling of tubulin, the neuropil marker synapsin, and several neuroactive substances (including histamine, serotonin, tyrosine hydroxylase, and orcokinin). Contrary to previous descriptions, the visual system displays a serial layout with only one first-order visual neuropil connected to a bilayered arcuate body by catecholaminergic interneurons. Fluorescent dye tracing reveals a previously reported second visual neuropil as the target of axons from the lateral sense organ instead of the eyes.ConclusionsGround pattern reconstruction reveals remarkable neuroanatomical stasis in the pycnogonid visual system since the Ordovician or even earlier. Its conserved layout exhibits similarities to the median eye pathway in euchelicerates, especially in xiphosurans, with which pycnogonids share two median eye pairs that differentiate consecutively during development and target one visual neuropil upstream of the arcuate body. Given multiple losses of median and/or lateral eyes in chelicerates, and the tightly linked reduction of visual processing centers, interconnections between median and lateral visual neuropils in xiphosurans and arachnopulmonates are critically discussed, representing a plausible ancestral condition of taxa that have retained both eye types.

Identifying the Key Components in ResNet-50 for Diabetic Retinopathy Grading from Fundus Images: A Systematic Investigation

Although deep learning based diabetic retinopathy (DR) classification methods typically benefit from well-designed architectures of convolutional neural networks, the training setting also has a non-negligible impact on the prediction performance. The training setting includes various interdependent components, such as objective function, data sampling strategy and data augmentation approach. To identify the key components in a standard deep learning framework (ResNet-50) for DR grading, we systematically analyze the impact of several major components. Extensive experiments are conducted on a publicly-available dataset EyePACS. We demonstrate that (1) the ResNet-50 framework for DR grading is sensitive to input resolution, objective function, and composition of data augmentation, (2) using mean square error as the loss function can effectively improve the performance with respect to a task-specific evaluation metric, namely the quadratically-weighted Kappa, (3) utilizing eye pairs boosts the performance of DR grading and (4) using data resampling to address the problem of imbalanced data distribution in EyePACS hurts the performance. Based on these observations and an optimal combination of the investigated components, our framework, without any specialized network design, achieves the state-of-the-art result (0.8631 for Kappa) on the EyePACS test set (a total of 42670 fundus images) with only image-level labels. Our codes and pre-trained model are available at https://github.com/YijinHuang/pytorch-classification

Batracobdelloides bangkhenensis sp. n. (Hirudinea: Rhynchobdellida), a new leech species parasite on freshwater snails from Thailand.

A new snail-eating leech, Batracobdelloides bangkhenensis sp. n., was discovered at Kasetsart University, Bangkhen Campus, Bangkok, Thailand. This species is found free living in the benthic zone of ponds; feeds on freshwater snails, including Bithynia siamensis siamensis, Indoplanorbis exustus, Radix rubiginosa, Physella acuta, and Pomacea canaliculata; and uses a shell as a shelter during the parental care period, with a colony of 7-15 juvenile individuals held on the venter inside the shell of host. Batracobdelloides bangkhenensis displays distinct morphological characters, including a rice-shaped body showing transparency, cephalization, two eye pairs merged on somite III, an anterior sucker twice as large as the cephalic region, a central mouth in the anterior sucker, seven light brown transverse rows in the neck region, absent dorsal papillae, rich green pigments on the dorsum, a male gonopore on XIIa2/XIIa3 (27-28), a female gonopore on XIIIa1/XIIIa2 (29-30), and diffuse aggregations of minute, spherical salivary glands in the neck region. Comparisons of the COI and COI-ND1 genes showed a monophyletic clade for Batracobdelloides, and the phylogenetic tree of the COI gene also indicated that B. bangkhenensis is distinct from other species in the genus, with strong support values.

A Method for Identification of Face of the Person and Presence of Facial Spot in the Face

Human Face and facial parts are the most significant parts as it reveals a person’s true identity. It plays an important role in various biometric applications like crowd analysis, human tracking, photography, cosmetic surgery, etc. There are many techniques are available to detect a facial image. Among them, skin detection is the most popular one. The aim of this paper is to detect first the person's identity from facial image and finally check any spot present the detected person. The first step is to detect the maximum skin region based on a combination method of RGB and HSV color space model. Next it is to verify the skin areas of human through machine learning approach. The Aggregated Channel Features (ACF) detector is used to identify the different facial parts like eye pairs, nose, and mouth. Bootstrap aggregation decision tree classifier is applied to classify the person’s identity based on Histogram Oriented Gradient (HOG) features value. The experimental results show that the proposed method gives the average 97% accuracy.

Fructosamine-3-Kinase as a Potential Treatment Option for Age-Related Macular Degeneration.

Age-related macular degeneration is the leading cause of blindness in the developed world. Since advanced glycation end products (AGEs) are implicated in the pathogenesis of AMD through various lines of evidence, we investigated the potential of fructosamine-3-kinase (FN3K) in the disruption of retinal AGEs, drusenoid material and drusenoid lesions in patients with AMD. AGE-type autofluorescence was measured to evaluate the effects of FN3K on glycolaldehyde-induced AGE-modified neural porcine retinas and unmodified human neural retinas. Eye pairs from cigarette-smoke- and air-exposed mice were treated and evaluated histologically. Automated optical image analysis of human tissue sections was performed to compare control- and FN3K-treated drusen and near-infrared (NIR) microspectroscopy was performed to examine biochemical differences. Optical coherence tomography (OCT) was used to evaluate the effect of FN3K on drusenoid deposits after treatment of post-mortem human eyes. FN3K treatment provoked a significant decrease (41%) of AGE-related autofluorescence in the AGE-modified porcine retinas. Furthermore, treatment of human neural retinas resulted in significant decreases of autofluorescence (−24%). FN3K-treated murine eyes showed less drusenoid material. Pairwise comparison of drusen on tissue sections revealed significant changes in color intensity after FN3K treatment. NIR microspectroscopy uncovered clear spectral differences in drusenoid material (Bruch’s membrane) and drusen after FN3K treatment. Ex vivo treatment strongly reduced size of subretinal drusenoid lesions on OCT imaging (up to 83%). In conclusion, our study demonstrated for the first time a potential role of FN3K in the disruption of AGE-related retinal autofluorescence, drusenoid material and drusenoid lesions in patients with AMD.

Profile of macular ganglion cell-inner plexiform layer thickness in healthy 6.5\u2009year- old Swedish children

BackgroundThe purpose was to study the macular ganglion cell- inner plexiform layer (GC-IPL) thickness in healthy 6.5 year- old Swedish children using Optical Coherence Tomography (OCT) and to study topography symmetry within eyes and between eye pairs.MethodsA total of 181 eyes of 92 healthy children (39 girls, 53 boys) aged 6.5 and serving as a term-born control group in the Extremely Preterm Infants in Sweden Study (EXPRESS), were examined with Cirrus HD-OCT. Main outcome measures were average and minimum values of GC-IPL thickness of the device’s predefined macular sectors. Single sectors, combined sectors defined as superior and inferior hemispheres and temporal and nasal sectors were evaluated. Intra-individual GC-IPL thickness between eye pairs was analyzed. Visual acuity, refraction and general cognition were assessed and correlated to GC-IPL outcome.ResultsEighty-five children completed the OCT examination and 155 out of 181 scans (86%) were analyzed. The mean average GC-IPL thickness was 85.9 μm (± 5.3; 5th and 95th percentiles were 76.0 and 94.6 μm). The mean minimum GC-IPL thickness was 83.6 μm (± 4.9; 5th and 95th percentiles were 75.4 and 92.3 μm). The difference in thickness between nasal and temporal sectors and between superior and inferior hemisphere sectors were less than 2 μm. The difference between average GC-IPL thickness and minimum GC-IPL thickness was 2.3 μm (± 1.9; 5th and 95th percentiles were 0.0 and 6.0 μm). The difference between the thickest and thinnest sector within eye was 6.4 μm (± 2.2; 5th and 95th percentiles were 3.0 and 10.0 μm). There was a moderate correlation in the difference between the nasal combined and the temporal combined sectors within eye pairs (p < 0.0001, Spearman’s ρ 0.58). The average GC-IPL thickness was weakly positively correlated with SE (spherical equivalent; combined sphere and ½ cylinder) (p = 0.031, Spearman’s ρ 0.23).ConclusionsThis study provides normative GC-IPL thickness values for healthy 6.5 year- old Swedish children. The GC-IPL thickness variations within eyes and within eye pairs are generally small. It could therefore be assumed that larger variations are sensitive markers of focal GC-IPL thinning due to damage to the primary visual pathways in children.

Endothelial Safety and Efficacy of Ex Vivo Collagen Cross-linking of Human Corneal Transplants

To assess endothelial safety and efficacy of exvivo corneal collagen cross-linking (CXL) in human corneal transplants stored in 2 different culture media. Fellow-eye controlled laboratory study of exvivo human donor corneas. Three sets of paired human donor corneas, 5 pairs each, were stored in organ culture medium before deswelling either at 31 C or at room temperature. One eye of each pair was cross-linked by 0.1% riboflavin in hydroxylpropyl methylcellulose (HPMC) instillation for 10minutes followed by 10minutes of ultraviolet-A (9 mW/cm2) irradiation while contralateral eyes served as controls. In Set 1, endothelial cell densities were determined. In Set 2, paired samples were assigned to the 2 deswelling media and CXL efficacy was assessed comparing to untreated controls using collagenase-A-assisted enzymatic digestion. In Set 3, biomechanical testing was performed in the eye pairs (treated vs control) by stress/strain measurements. There was no difference in endothelial cell counts between CXL samples and controls (P= .21). No statistically significant difference in digestion dynamics was found between tissues stored in the 2 different culture media. Complete enzymatic digestion was slowed down by 3 hours in the cross-linked samples (P= .036). Stress needed for a 12% strain was increased by 34% in the treatment group compared to control (P= .04). Exvivo CXL of human donor tissue is an effective and safe procedure with no difference regarding efficacy between 2 commercially available deswelling media. Biochemical and biomechanical resistance were significantly increased after CXL. Patients requiring keratoplasty owing to corneal melting might benefit from the strengthening effect of preoperative CXL of donor tissue.

Detection of Iris Localization in Facial Images Using Haar Cascade Circular Hough Transform

In numerous science applications, face detection and iris extraction have been recognized as crucial stages by getting more consideration among researchers as it has an important job. This paper presents an automatic detection method of the iris and its center detection by applying the Haar Cascade Classifier and the Circular Hough Transform algorithm. The suggested method is divided into two primary methodologies: face recognition utilizes the Haar Cascade Classifier and iris extraction using the Hough Transform. The system detects the face from a set of facial images using an Impa-faced dataset. The improved AdaBoost algorithm constructs a cascaded classifier for face detection. Then, by applying the Haar Cascade to obtain an eye pair region and a Hough transform for iris detection by extracting Haar features. Finally, the improved circular Hough transform algorithm locates the iris center. The experimental results of the suggested method show a high-speed, robust ability to acquire the coordinates of the iris center accurately under various illumination changes on different states of human images. The overall accuracy for locating the iris center was 98.75%.

Eye center localization in a facial image based on geometric shapes of iris and eyelid under natural variability

The estimation of the human eye centers is one of the important step in several computer vision applications such as driver drowsiness detection, eye tracking, face recognition etc. Most of the existing techniques are able to localize eyes in frontal faces only while they fail to localize eye pairs in complex scenarios such as changes in head pose, scale, and illumination. In this paper, an eye localization method has been proposed that can locate the eye centers more precisely in facial images captured under the above-mentioned complexities. The proposed method consists of three stages: eye candidate detection, eye candidate verification, and post-processing. In eye candidate detection, the possible eye candidates are extracted using two new features namely Semi-Circular Edge Shape (sCES) and Semi-Ellipse Edge Shape (sEES) features. These features take into consideration the semi-circular and semi-ellipse edges of iris and eyelid and hence are able to localize eye centers more precisely. In verification, the extracted eye candidates are verified using a Support Vector Machine (SVM) based classifier. A scale-space framework is also included in the verification stage to handle the scale variations of images. In post-processing, the eye centers are paired using some geometrical constraints and then a modified gradient-based method is proposed to detect the required eye pair. The proposed system is evaluated on different databases to check its robustness to changes in head pose, scale, illumination etc. The experimental results suggest that the proposed method shows better accuracy in challenging environments and also outperforms some state-of-the-art methods.

The visual spectral sensitivity of the Chilean recluse spider Loxosceles laeta.

Spiders are a large group of arthropods and nearly omnipresent in warm and temperate climates. They rely on tactile and visual information to hunt and breed, but compared with their mechanical senses, little is known about their visual systems. In this study, we analyzed the visual spectral sensitivity of the Chilean recluse spider Loxosceles laeta, a synanthropic species posing a significant threat to humans, using electroretinogram recordings of its three eye types and open field tests with localized chromatic illumination for behavioral analysis. The electroretinogram displayed two sensitivity peaks in the ultraviolet and green ranges, and no differences were observed between the three eye types and between male and female specimens. Selective chromatic adaptation reduced overall light sensitivity, but did not support the expression of more than one type of rhodopsin in photoreceptors. The open field tests revealed a preference for corners over side areas, and an increased exploration of open field areas illuminated by shorter wavelength (violet to green) light compared with non-illuminated areas, while no behavioral responses to red and near-infrared light were observed. These data suggest that L. laeta has monochromatic vision without spectral specializations in its three secondary eye pairs.