Natural Light Conditions Research Articles

Vision-based normalized canopy area estimation for variable nitrogen application in apple orchards

Normalized canopy area is a measure of spatial canopy fill or canopy coverage in a tree. It is an important factor for growers to consider in making various farming decisions, such as timing and amount of fertilization, pesticide application, and irrigation, as it gives a sense of the overall canopy growth. This study focused on estimating the normalized canopy area of apple trees using a ground-based stereo-vision system in a commercial orchard under natural lighting conditions. A color, depth, and height threshold, followed by a point density-based outlier removal was applied to segment the target tree canopies. The segmented canopy area was compared with the area estimated with manual segmentation, which showed an F1 score of 0.78. The overall growth of the tree canopies was then represented using normalized canopy area, which was calculated as a fraction of pixels occupied by the tree foliage over the total possible area a tree could occupy for a given tree spacing. It was found that the normalized canopy area of individual trees was highly correlated with the surface volume of the canopy (r = 0.7), which is a traditionally used parameter to assess or represent canopy growth. The estimated normalized canopy area was then compared against experts' nitrogen recommendation for the sample trees, which was based on their visual assessment of the vigor/growth of the trees. Four individual experts (one apple farmer, one orchard manager, and two horticulturists with research and outreach experience) were asked to provide their independent nitrogen recommendation and an average recommendation level was calculated. The comparison showed an overall correlation coefficient of -0.5, indicating an opposite relationship (higher the canopy density, lower the need for nitrogen and vice versa) between the normalized canopy area and the average of the experts' nitrogen recommendation. When the relationship between the normalized area and nitrogen recommendation level was assessed for individual experts, the correlation coefficient was estimated to be -0.86, -0.84, -0.96, and -0.78, respectively, for Expert1, Expert2, Expert3, and Expert4 respectively. The results from this study identified a critical measure for assessing nitrogen needs at individual tree levels in apple orchards and provides an automated tool to estimate the measure. The outcomes of this study could be used as a part of an integrated decision support system for orchard operations (e.g., fertilization or irrigation) that may include additional canopy characteristics such as spectral signature, trunk diameter, and rate of change of leaf color in the fall.

Construction of novel bimetallic Ti/Ce-MOFs for ratiometric fluorescence sensing of trace copper and photocatalytic reduction of chromium (Ⅵ)

A new bimetallic-centered MOFs, NH2-MIL-125(Ti/Ce), was synthesized by partially replacing Ti with Ce. The doping of Ce endowed the MOFs oxidase-like activity and enabled the construction of a ratiometric fluorescence sensing system in combination with o-phenylenediamine (OPD) for detecting trace Cu2+ with a limit of detection of 16 nM and a linear range of 0.02–50 μM. Trace Cu2+ was successfully quantified in real water samples. The doping of Ce also increased the photocatalytic capability of the MOFs by about 2 times, achieving excellent reduction of Cr(Ⅵ) under natural light conditions.

Presence of the cloud cover and elevation angle of the sun affect measurements of eggshell coloration and patterning obtained from calibrated digital images

AbstractCalibrated digital photography is frequently used in studies focusing on avian eggshell appearance to measure colour and pattern features. Photographs are often taken in natural light conditions, yet little is known to what extent the normalisation process is able to control for varied light. Here, we photographed 36 blown eggs of the Japanese quail Coturnix japonica at five different elevation angles of the sun on both sunny and uniformly overcast days alongside grey standards. We normalised and processed the photographs in the MICA Toolbox software and checked how much noise was introduced by different natural light conditions to the colour and pattern measurements of the same set of eggs. Our results indicate that natural variation of light conditions affects eggshell colour and pattern measurements obtained by means of calibrated digital photography. Depending on a trait, the elevation angle of the sun had similar or even greater effect on the measurement than the presence of the cloud cover. Furthermore, measurements taken in cloudy conditions were more repeatable than those taken in sunny conditions. Based on the results, we propose practical guidelines regarding measuring colour and pattern of eggshells using calibrated digital photography in outdoor conditions.

Comparative study of transcriptome profiles in the eyes and brain of the tiger puffer Takifugu rubripes brooder under red light filtered LED

Light spectrum is an important environmental factor affecting fish activities. However, the existing studies on spectrum are usually conducted under monochromatic light, which are multiple variables compared to the natural light conditions and increase uncertainty. The results may be caused by the absence of all other chromatic lights. The tiger puffer Takifugu rubripes is an important commercial marine fish in China, Japan, and Korea. In this study, the effects of red light on the eyes and brain of T. rubripes brooder, including histological changes in retina and optic tectum, the levels of oxidative stress indicators in serum, and changes in gene expression at transcriptome level were examined, after the treatment of full spectrum LED (Fs) and red light filtered LED (Rf) for 52 days. The results showed that the growth and histological structure of retina and optic tectum were not significantly altered. While, Rf significantly decreased the activity of superoxide dismutase and catalase, and increased the content of cortisol (P < 0.05), but had no effect on the levels of glutathione peroxidase and malondialdehyde. Transcriptome data revealed that there were 147 differentially expressed genes (DEGs) in the eyes of Rf vs Fs. Many DEGs related to vision changed significantly. KEGG enrichment and GSEA analyses showed that the significantly enriched arachidonic acid pathway down-regulated in the Rf group, in which genes related to sexual maturation and ovulation presented significant changes, suggesting that Rf may affect the ovulation process through this pathway. In the brain, the number of DEGs was only 20, which are most correlated with circadian rhythm. The present study displayed for the first time the role of red light in promoting ovulation in mature fish by filtering it, which may provide more supports to highlight the function of red light in fish culture.

Steroidogenic potential of the gonad during sex differentiation in the rice field frog Hoplobatrachus rugulosus (Anura: Dicroglossidae).

Prior studies demonstrated that gonadal differentiation in the rice field frog, Hoplobatrachus rugulosus, was of an undifferentiated type since all individuals had ovaries at complete metamorphosis. However, the steroidogenic potential of the gonad is still unknown. In this study, H. rugulosus were obtained by stimulating fertilization in the laboratory under natural light and temperature conditions. The gonads were collected and their steroidogenic potential was evaluated by determining the expression level of messenger RNA (mRNA) encoding for cytochrome P450 17-hydroxylase/C17-20 lyase (CYP17) and cytochrome P450 aromatase (CYP19) using quantitative real-time RT-PCR and the localization of CYP17 mRNA in tissues by in situ hybridization. The CYP17 mRNA levels in males at 4-11 weeks postmetamorphosis were higher than in female and intersex gonads. This corresponded to their localization in the gonadal tissues, where CYP17 signals were specifically detected in the Leydig cells of the testis at 5-16 weeks postmetamorphosis but was undetectable in all ovary samples. The CYP19 mRNA levels in females at 4-11 weeks postmetamorphosis was higher than in male and intersex gonads, which corresponded with gonadal development, indicating the potential steroidogenic function of the ovary. Based on the present results, the role of CYP17 and CYP19 mRNA in sex differentiation in H. rugulosus may occur after gonadal sex differentiation and the steroidogenic potential of the gonads exhibited a sexual dimorphic pattern. These results provide a crucial basis for further research on the developmental biology in anuran species.

Shading-Dependent Greening Process of the Leaves in the Light-Sensitive Albino Tea Plant 'Huangjinya': Possible Involvement of the Light-Harvesting Complex II Subunit of Photosystem II in the Phenotypic Characteristic.

The light-sensitive albino tea plant can produce pale-yellow shoots with high levels of amino acids which are suitable to process high-quality tea. In order to understand the mechanism of the albino phenotype formation, the changes in the physio-chemical characteristics, chloroplast ultrastructure, chlorophyll-binding proteins, and the relevant gene expression were comprehensively investigated in the leaves of the light-sensitive albino cultivar 'Huangjinya' ('HJY') during short-term shading treatment. In the content of photosynthetic pigments, the ultrastructure of the chloroplast, and parameters of the photosynthesis in the leaves of 'HJY' could be gradually normalized along with the extension of the shading time, resulting in the leaf color transformed from pale yellow to green. BN-PAGE and SDS-PAGE revealed that function restoration of the photosynthetic apparatus was attributed to the proper formation of the pigment-protein complexes on the thylakoid membrane that benefited from the increased levels of the LHCII subunits in the shaded leaves of 'HJY', indicating the low level of LHCII subunits, especially the lack of the Lhcb1 might be responsible for the albino phenotype of the 'HJY' under natural light condition. The deficiency of the Lhcb1 was mainly subject to the strongly suppressed expression of the Lhcb1.x which might be modulated by the chloroplast retrograde signaling pathway GUN1 (GENOMES UNCOUPLED 1)-PTM (PHD type transcription factor with transmembrane domains)-ABI4 (ABSCISIC ACID INSENSITIVE 4).

Natural vs. Artificial Light: A Study on the Influence of Light Source on Chlorophyll Content and Photosynthetic Rates on Indoor Plants

Indoor landscaping is an environmentally friendly approach that enriches the environment and fosters productivity and comfort for occupants. The practice of incorporating plants into interior spaces requires meticulous care to ensure healthy growth and prolong the benefits of interior greening. This study explores the impact of natural and artificial light, represented by fluorescent lighting on the growth and physiological responses of Codiaeum variegatum and Ardisia japonica. A natural light chamber and an artificial light chamber of identical dimensions were constructed to compare the plants’ physiological responses under consistent temperature, humidity, and illuminance conditions. The results indicate that Codiaeum variegatum and Ardisia japonica exhibited higher chlorophyll content and photosynthetic rates under natural light conditions compared to fluorescent lighting. Furthermore, the study found that natural light offers a rich spectral distribution across various wavelengths, providing an advantage for plant growth. Although direct comparisons between natural and artificial light environments are inherently challenging due to the distinct characteristics of each light source, the study emphasizes the importance of considering the rich spectral distribution of natural light when designing artificial lighting systems for optimal plant growth. In conclusion, understanding the effects of natural and artificial light on indoor plants is crucial to supporting plant growth and creating more effective indoor gardening solutions. Although direct comparisons between natural and artificial light environments are inherently challenging due to the distinct characteristics of each light source, natural light provides a more advantageous environment for growth compared to fluorescent lighting, with Codiaeum variegatum and Ardisia japonica both exhibiting a higher chlorophyll content and photosynthetic rate under natural light conditions.

The change in metabolic activity of a large benthic foraminifera as a function of light supply

We studied metabolic activity of the symbiont-bearing large benthic foraminifer Heterostegina depressa under different light conditions. Besides the overall photosynthetic performance of the photosymbionts estimated by means of variable fluorescence, the isotope uptake (13C and 15N) of the specimens (= holobionts) was measured. Heterostegina depressa was either incubated in darkness over a period of 15 days or exposed to an 16:8 h light:dark cycle mimicking natural light conditions. We found photosynthetic performance to be highly related to light supply. The photosymbionts, however, survived prolonged darkness and could be reactivated after 15 days of darkness. The same pattern was found in the isotope uptake of the holobionts. Based on these results, we propose that 13C-carbonate and 15N-nitrate assimilation is mainly controlled by the photosymbionts, whereas 15N-ammonium and 13C-glucose utilization is regulated by both, the symbiont and the host cells.

FsHemF is involved in the formation of yellow Forsythia leaves by regulating chlorophyll synthesis in response to light intensity

The leaves of Forsythia koreana ‘Suwon Gold’ are yellow under natural light condition and can revert to green when the light intensity is reduced. To understand the molecular mechanism of leaf color changes in response to light intensity, we compared the chlorophyll content and precursor content between yellow- and green-leaf Forsythia under shade and light-recovery conditions. We identified the conversion of coproporphyrin III (Coprogen III) to protoporphyrin IX (Proto IX) as the primary rate-limiting step of chlorophyll biosynthesis in yellow-leaf Forsythia. Further analysis of the activity of the enzymes that catalyze this step and the expression pattern of the chlorophyll biosynthesis-related genes under different light intensities revealed that the negatively regulated expression of FsHemF by light intensity was the major cause affecting the leaf color change in response to light intensity in yellow-leaf Forsythia. To further understand the cause of differential expression pattern of FsHemF in yellow- and green-leaf lines, we compared the coding sequence and promoter sequence of FsHemF between yellow- and green-leaf Forsythia. We found that one G-box light-responsive cis-element was absent in the promoter region of green-leaf lines. To investigate the functional role of FsHemF, we performed virus-induced gene silencing (VIGS) of FsHemF in green-leaf Forsythia, which leads to yellowing leaf veins, decreased chlorophyll b content, and inhibition of chlorophyll biosynthesis. The results will assist in elucidating the mechanism of yellow-leaf Forsythia in response to light intensity.

Determining Regression Models for Photosynthesis and Stomatal Resistance as Affected by Temperature and Light Intensity in Tomato (Lycopersicon Esculentum Mill.) and Eggplant (Solanum Melongena L.) Grown in Glasshouses

This study was carried out to examine the relationships between net leaf photosynthesis and temperature and light intensity, between stomatal resistance and temperature and light intensity in tomato and aubergine grown with a range temperature from 10 to 30 °C and different light intensities from 3 to 7 MJm-2 d-1. The study was carried out in a six-compartment greenhouse (size 4 m * 8 m), the temperature of which can be controlled by air conditioning, on tomato and eggplant plants. Each of the six greenhouse compartments was set to have maximum temperatures of 10, 12, 16, 18, 20 and 24 °C. Commercial varieties named "Counter" for tomato and "Bonica" for eggplant were used. "Fisons M2" commercial compost was used in all growing media and nutrient was applied equally. In the study, different sowing and planting dates were applied to benefit from natural light conditions (between 3 and 7 MJm-2d-1). Average temperature in each compartment was recorded using a 'Combine' data logger at 15 minute intervals. A porometer (Delta-T device, MT -3) was used to measure the stomatal resistance of tomato and eggplant leaves. The stomatal resistance measurements of the plants were made at the same time of the day (between 11.00-13.00) at 15-day intervals at the top, middle and lower levels of the crown of four different plants in different environmental conditions. In tomato, leaf photosynthesis increased curvilinearly with temperatures up to about 20.5 °C at low light intensity and declined at higher temperatures. The highest photosynthesis was obtained from the plants grown at a temperature of 22.5 °C and 7 MJm-2d-1 light intensity. The lowest photosynthesis was at 10 °C and 3 MJm-2d-1. In aubergine, at low light intensities, net photosynthesis increased curvilinearly up to 23 °C while it increased up to 20 °C at high light intensities and declined at higher temperatures. Maximum net leaf photosynthesis was found to be greater in tomato than aubergine.

3D SCENE RECONSTRUCTION AND DIGITIZATION METHOD FOR MIXED REALITY SYSTEMS

Mixed reality systems are a promising direction of research that opens up great opportunities forinteraction with virtual objects in the real world. Like any promising direction, mixed reality has a number of unresolved problems. One of these problems is the synthesis of natural lighting conditions for virtual objects, including the correct light interaction of virtual objects with the real world. Since virtual and real objects are located in different spaces, it is difficult to ensure their correct interaction. To create digital copies of realworld objects, machine learning tools and neural network technologies are employed. These methods are successfully used in computer vision for space orientation and environment reconstruction. As a solution, it is proposed to transfer all objects into the same information space: virtual space. This makes it possible to solve most of the problems associated with visual discomfort caused by the unnatural light interaction of real and virtual objects. Thus, the basic idea of the method is to recognize physical objects from point clouds and replace these objects with virtual CAD models. In other words, it implies semantic analysis of a scene and classification of objects with their subsequent transformation into polygonal models. In this study, we use competitive neural network architectures, which can provide state-of-the-art results. The test experiments are carried out on Semantic3D, ScanNet, and S3DIS, which are currently the largest datasets with point clouds that represent indoor scenes. For semantic segmentation and classification of 3D point clouds, we use the PointNeXt architecture based on PointNet, as well as modern methods of data augmentation in the process of learning. For geometry reconstruction, the Soft Rasterizer differentiable rendering method and the Total3Understanding neural network are considered.

Light conditions during Atlantic salmon embryogenesis affect key neuropeptides in the melanocortin system during transition from endogenous to exogenous feeding.

During the first feeding period, fish will adapt to exogenous feeding as their endogenous source of nutrients is depleted. This requires the development of a functional physiological system to control active search for food, appetite, and food intake. The Atlantic salmon (Salmo salar) melanocortin system, a key player in appetite control, includes neuronal circuits expressing neuropeptide y (npya), agouti-related peptide (agrp1), cocaine- and amphetamine-regulated transcript (cart), and proopiomelanocortin (pomca). Little is known about the ontogeny and function of the melanocortin system during early developmental stages. Atlantic salmon [0-730 day degrees (dd)] were reared under three different light conditions (DD, continuous darkness; LD, 14:10 Light: Dark; LL, continuous light) before the light was switched to LD and the fish fed twice a day. We examined the effects of different light conditions (DD LD , LD LD , and LL LD ) on salmon growth, yolk utilization, and periprandial responses of the neuropeptides npya1, npya2, agrp1, cart2a, cart2b, cart4, pomca1, and pomca2. Fish were collected 1 week (alevins, 830 dd, still containing yolk sac) and 3 weeks (fry, 991 dd, yolk sac fully consumed) into the first feeding period and sampled before (-1 h) and after (0.5, 1.5, 3, and 6 h) the first meal of the day. Atlantic salmon reared under DD LD , LD LD , and LL LD had similar standard lengths and myotome heights at the onset of first feeding. However, salmon kept under a constant light condition during endogenous feeding (DD LD and LL LD ) had less yolk at first feeding. At 830 dd none of the neuropeptides analyzed displayed a periprandial response. But 2 weeks later, and with no yolk remaining, significant periprandial changes were observed for npya1, pomca1, and pomca2, but only in the LD LD fish. This suggests that these key neuropeptides serve an important role in controlling feeding once Atlantic salmon need to rely entirely on active search and ingestion of exogenous food. Moreover, light conditions during early development did not affect the size of salmon at first feeding but did affect the mRNA levels of npya1, pomca1, and pomca2 in the brain indicating that mimicking natural light conditions (LD LD ) better stimulates appetite control.

Color induction in the restoration of architecture in historic city centers

AbstractIn the restoration of historic facades, specialists need to tackle the problem of colors applied to the ornamental elements that appear different on a facade than if they were isolated, because the colors of the surrounding surfaces influence them through color induction. This study evaluates the difference between the colors of ornaments observed in isolation, which we call nominal colors, and on different background colors, which we call induced colors, considering the historic color palette of Valencia as studied by the Color Research Group in Architecture of the Universitat Politècnica de València. The study consisted of the evaluation of 108 color pairs painted on physical cardboards, rated by 26 female students (average age 26.5), under natural lighting conditions, and using the Natural Color System (NCS) Color Atlas for the color match. Results indicate that induced colors have lower NCS blackness compared with nominal ones, and slightly higher or no NCS chromaticness shift. Nominal colors with low chromaticness exhibit an important hue induction effect and tend to shift to a hue that is located opposite to that of the background color in the NCS color circle, this effect being larger when the background has a warm color. The induction effect is larger when ornamental elements and background have similar nominal NCS blackness, and different nominal NCS chromaticness,with nominal hue difference no significant in the induction effect. The colors of ornamental elements that entailed greater induction were those with the lowest NCS chromaticness, and in the case of the background color, those with the highest chromaticness.

The concept of aerial photography using a two-element active optoelectronic complex

The article proposes the concept of obtaining images on the basis of which it is possible to create three-dimensional models of objects. In particular, images contain the data necessary to recreate the three-dimensional form of an object. Such data include: the length, width of the modeled object as well as the lengthening of the shadow cast on the base surface. In accordance with the proposed concept, an important condition for obtaining images suitable for interpretation is the use of spatially separated equipment generating and recording optical radiation. The orientation parameters of the illumination source and the camera are selected, taking into account the requirements for the photogrammetric quality of images. A characteristic of geometric distortions that occur when changing the shooting mode from plan to perspective is presented, which demonstrates the change in the aspect ratio of the image depending on the angle of the camera. A characteristic of the shadow elongation depending on the spatial position of the optical radiation source is presented, which shows the influence of the orientation parameters of the optical illumination means on the length of the shadow cast on the base surface. On the basis of these characteristics, the choice of parameters of relative orientation in space of the source of optical illumination and the equipment that detects optical radiation is substantiated. Predicting the value of geometric distortions of images at the stage of choosing the parameters of relative orientation of the equipment of a two-element active optoelectronic complex allows you to save the photogrammetric quality of images and, as a result, measure the length and width of an object. Predicting the magnitude of the shadow elongation under conditions of artificial optical illumination provides the possibility of transferring features in the image for calculating the applicate and, as a result, recreating the three-dimensional shape of the object. The proposed concept of image registration finds application in topographic and geodetic and engineering work in conditions of insufficient natural light. For example, the use of a two-element active optical-electronic complex makes it possible to obtain three-dimensional photo plans of the terrain of geographic regions with a short daylight hours.

A comparison of consistent UV treatment versus inconsistent UV treatment in horticultural production of lettuce

UV radiation is an underrated radiation currently missing in many horticultural production systems of vegetables in protected cultivation. It can be added e.g., in LED light sources. Using lettuce as a model plant, this study determined whether the use of UVB LEDs is suitable (1) for use in consistent systems (indoor farming) or (2) inconsistent systems (greenhouse). Blue and red LEDs were selected as additional artificial lighting to UVB LEDs. Both approaches led to a reproducible increase of desired flavonol glycosides, such as quercetin-3-O-(6′′-O-malonyl)-glucoside or quercetin-3-O-glucuronide and the anthocyanin cyanidin-3-O-(6′′-O-malonyl)-glucoside in lettuce. The impact of the consistent UVB treatment is higher with up to tenfold changes than that of the inconsistent UVB treatment in the greenhouse. Varying natural light and temperature conditions in greenhouses might affect the efficiency of the artificial UVB treatment. Here, UVB LEDs have been tested and can be recommended for further development of lighting systems in indoor farming and greenhouse approaches.

Documenting Outdoor Simulated Scenes with Photogrammetry

In both traditional and forensic archaeology, the thorough documentation of context is essential to a proper excavation or recovery, as the processing of archaeological sites and forensic scenes is an inherently destructive process. Over the past few decades, close-range photogrammetry (CRP) has become increasingly utilized by archaeologists to digitally record their excavations. However, CRP has not been employed regularly during outdoor forensic archaeological scene recoveries, and therefore protocols are not well defined for this context. One important consideration when employing CRP is the impact of natural lighting conditions. Light is an important variable when recording scenes using CRP because the quality of light can affect the visual quality of the model and imagery. In particular, shadows from the photographer’s body, trees, and other structures are unavoidable when collecting images around the perimeter of scenes, a requirement of photogrammetry data collection. Therefore, the purpose of the present study was to determine whether implementing light correction tools would improve the three-dimensional model imagery. Two mock forensic scenes with different lighting conditions were photographed four times each in order to test the application of light correction tools, including a large tarp and artificial lights, for mitigating harsh lighting conditions. The results demonstrate that a tarp is a viable light correction tool that can improve the visual quality of the final models by eliminating lighting inconsistencies in both open and wooded environments. Based on these results, improved guidelines for the application of CRP to outdoor forensic archaeological scenes are presented.

Low Image Contrast Detection in a Bright Light Interference HDR Scene Using Smart CAOS Camera

Proposed and experimentally demonstrated is the linear low contrast detection capability of a smart design Coded Access Optical Sensor (CAOS) camera operating under interfering bright light conditions. This fine contrast imaging capability is delivered by engaging both the silicon CMOS sensor-mode of the constructed smart CAOS camera and deploying spatial filtering within its inherent on/off light routing structure via the Digital Micromirror Device (DMD). The camera uses the CMOS-mode identified non-bright light region of interest to engage the Code-Division Multiple Access (CDMA) CAOS-mode to achieve successful target imaging. Specifically, the low contrast variation (¡ 6 dB dynamic range) test target region of the toy horse target is successfully experimentally imaged within a near 87 dB High Dynamic Range (HDR) scene created by the presence of an external interfering bright visible light source. Applications scenarios that can benefit from the use of the proposed camera include natural bright light conditions surveillance as well as bright laser and LED targeted imaging across medical, industrial and scientific sectors.

WE3DS: An RGB-D Image Dataset for Semantic Segmentation in Agriculture

Smart farming (SF) applications rely on robust and accurate computer vision systems. An important computer vision task in agriculture is semantic segmentation, which aims to classify each pixel of an image and can be used for selective weed removal. State-of-the-art implementations use convolutional neural networks (CNN) that are trained on large image datasets. In agriculture, publicly available RGB image datasets are scarce and often lack detailed ground-truth information. In contrast to agriculture, other research areas feature RGB-D datasets that combine color (RGB) with additional distance (D) information. Such results show that including distance as an additional modality can improve model performance further. Therefore, we introduce WE3DS as the first RGB-D image dataset for multi-class plant species semantic segmentation in crop farming. It contains 2568 RGB-D images (color image and distance map) and corresponding hand-annotated ground-truth masks. Images were taken under natural light conditions using an RGB-D sensor consisting of two RGB cameras in a stereo setup. Further, we provide a benchmark for RGB-D semantic segmentation on the WE3DS dataset and compare it with a solely RGB-based model. Our trained models achieve up to 70.7% mean Intersection over Union (mIoU) for discriminating between soil, seven crop species, and ten weed species. Finally, our work confirms the finding that additional distance information improves segmentation quality.

Study on long-term morphological stability of three-dimensional-printed photosensitive resin dental models

Objective: To study the long-term morphological stability of three-dimensional (3D) printed photosensitive resin dental models under natural light and dark conditions. Methods: Eighty sets of resin dental models were made by the desktop 3D printer from one digital standard model set, and randomly divided into two groups, namely natural light group (40 sets) and dark group (40 sets). All resin models were stored in sealed bags, with 4 model sets from each group randomly collected after 1, 3, 5, 7, 14, 21, 28, 40, 60, or 90 days of storage and 3D scanned using an optical model scanner. The root-mean-square error (RMSE) was calculated to represent the mean deviation of the difference between the digital standard model and the scanned resin model. Meanwhile, three linear indexes (the width between the canines, the width between the first molars, and the arch length) of the resin dental model were measured and compared with the corresponding values of the standard model. RMSE and the linear measurements between the digital standard model and the scanned resin models were compared between the natural light group and the dark group and among models from different time points. Results: Compared with the digital standard model, the RMSE values of 96.9% (155/160) resin dental models were less than 0.1 mm within 90-day storage. Also, at the same time point, there was no significant difference in the RMSE between the natural light group and the dark group (P>0.05). 75.0% (360/480) of the absolute values of the linear differences (differences in inter-canine width, intra-molar width, and arch length between the digital standard model and the scanned resin model) were within 0.2 mm, and about 0.1% (3/480) of the linear differences were greater than 0.5 mm, and all of the linear differences were within 0.6 mm. Conclusions: 3D-printed resin dental models can be stored stably under natural light and dark conditions for a long time.

Underwater Optical Image Restoration Method for Natural/Artificial Light

This paper investigates the underwater optical image restoration method under the background of underwater target detection based on optical vision in AUVs. The light source used for AUV detection is different when the AUV operates in different depths. The natural light source is used in shallow water and the artificial light source is used in deep water. This paper investigates underwater optical image restoration in these two light conditions. Aiming at the problem of image blurring in underwater optical images, the traditional underwater image restoration method based on scattering model can obtain satisfactory image restoration performance in natural light conditions. However, it cannot obtain the same image restoration result in artificial light conditions. To solve this problem, this paper presents an improved underwater optical image restoration method based on the scattering model. The scattering model and power spectrum are used to solve the initial parameters of the filter, and the parameters are optimized based on an evaluation index. The index of image definition is introduced to evaluate the restoration performance and to achieve the satisfactory image restoration result in both natural light and artificial light conditions. The effectiveness of the presented method is verified by experiments.