Integrating multi-spectral imaging and Raman spectroscopy for in vivo endoscopic assessment of rat intestinal tract.

An integrated system for in vivo multi-spectral imaging (MSI) and Raman spectroscopy was developed to understand the external morphology and internal molecular information of biological tissues. The achieved MSI images were reconstructed by eighteen separated images from 400 nm to 760 nm, whose illumination bands were selected with six tri-channel band filters. Based on the spectral analysis algorithms, the spatial distribution patterns of blood volume, blood oxygen content and tissue scatterer volume fraction were visualized. In vivo Raman spectral measurements were executed by inserting specially designed optical probe into instrumental channel of endoscope. By this way, the molecular composition at selected sampling points could be identified with its fingerprint spectral information under the guidance of molecular imaging modality. Therefore, both structural and compositional features of intestinal membrane could be addressed without labeling and continuously. The achieved results testified that our presented methodology reveals insights not easily extracted from either MSI or Raman spectroscopy individually, which brings the enrichment of biological and chemical meanings for future in vivo studies.

An Evaluation of Multi-Channel Sensors and Density Estimation Learning for Detecting Fire Blight Disease in Pear Orchards.

Fire blight is an infectious disease found in apple and pear orchards. While managing the disease is critical to maintaining orchard health, identifying symptoms early is a challenging task which requires trained expert personnel. This paper presents an inspection technique that targets individual symptoms via deep learning and density estimation. We evaluate the effects of including multi-spectral sensors in the model's pipeline. Results show that adding near infrared (NIR) channels can help improve prediction performance and that density estimation can detect possible symptoms when severity is in the mid-high range.

Toward noncontact macroscopic imaging of multiple cancers using multi-spectral inelastic scattering detection.

Here we introduce a Raman spectroscopy approach combining multi-spectral imaging and a new fluorescence background subtraction technique to image individual Raman peaks in less than 5 seconds over a square field-of-view of 1-centimeter sides with 350 micrometers resolution. First, human data is presented supporting the feasibility of achieving cancer detection with high sensitivity and specificity - in brain, breast, lung, and ovarian/endometrium tissue - using no more than three biochemically interpretable biomarkers associated with the inelastic scattering signal from specific Raman peaks. Second, a proof-of-principle study in biological tissue is presented demonstrating the feasibility of detecting a single Raman band - here the CH2/CH3 deformation bands from proteins and lipids - using a conventional multi-spectral imaging system in combination with the new background removal method. This study paves the way for the development of a new Raman imaging technique that is rapid, label-free, and wide field.

Advanced Techniques for Liver Fibrosis Detection: Spectral Photoacoustic Imaging and Superpixel Photoacoustic Unmixing Analysis for Collagen Tracking.

Liver fibrosis, a major global health issue, is marked by excessive collagen deposition that impairs liver function. Noninvasive methods for the direct visualization of collagen content are crucial for the early detection and monitoring of fibrosis progression. This study investigates the potential of spectral photoacoustic imaging (sPAI) to monitor collagen development in liver fibrosis. Utilizing a novel data-driven superpixel photoacoustic unmixing (SPAX) framework, we aimed to distinguish collagen presence and evaluate its correlation with fibrosis progression. We employed an established diethylnitrosamine (DEN) model in rats to study liver fibrosis over various time points. Our results revealed a significant correlation between increased collagen photoacoustic signal intensity and advanced fibrosis stages. Collagen abundance maps displayed dynamic changes throughout fibrosis progression. These findings underscore the potential of sPAI for the noninvasive monitoring of collagen dynamics and fibrosis severity assessment. This research advances the development of noninvasive diagnostic tools and personalized management strategies for liver fibrosis.

Image capturing, segmentation and data analysis of shredded refuse streams.

Refuse sorting is an important cornerstone of the recycling industry, but ever-changing refuse compositions and the desire to increase recycling rates still pose many unsolved challenges. The digitalisation of refuse sorting plants promises to overcome these challenges by optimising and automatically adapting the sorting process. This publication describes a system for image capturing, segmentation-based refuse recognition and data analysis of shredded refuse streams. The image capturing collects multispectral 2D and 3D images of the refuse streams on conveyor belts. The image recognition performs a semantic segmentation of the images to determine the refuse composition from the 2D images, whereas the 3D images approximate the volumes on the conveyor belts. The semantic segmentation is done by a combined convolutional neural network model, consisting of a foreground-background and a refuse class segmentation. Both models rely on synthetic training data to reduce the necessary amount of manually labelled training data, whereas the final segmentation performance reaches an Intersection over Union of up to 75%. The results of the semantic segmentation and volume estimation are combined with data of the shredding machinery by transforming it into a unified representation. This combined dataset is the basis for estimating the processed refuse masses from the semantic segmentation and volume estimation.

LodgeNet: an automated framework for precise detection and classification of wheat lodging severity levels in precision farming.

Wheat lodging is a serious problem affecting grain yield, plant health, and grain quality. Addressing the lodging issue in wheat is a desirable task in breeding programs. Precise detection of lodging levels during wheat screening can aid in selecting lines with resistance to lodging. Traditional approaches to phenotype lodging rely on manual data collection from field plots, which are slow and laborious, and can introduce errors and bias. This paper presents a framework called 'LodgeNet,' that facilitates wheat lodging detection. Using Unmanned Aerial Vehicles (UAVs) and Deep Learning (DL), LodgeNet improves traditional methods of detecting lodging with more precision and efficiency. Using a dataset of 2000 multi-spectral images of wheat plots, we have developed a novel image registration technique that aligns the different bands of multi-spectral images. This approach allows the creation of comprehensive RGB images, enhancing the detection and classification of wheat lodging. We have employed advanced image enhancement techniques to improve image quality, highlighting the important features of wheat lodging detection. We combined three color enhancement transformations into two presets for image refinement. The first preset, 'Haze & Gamma Adjustment,' minimize atmospheric haze and adjusts the gamma, while the second, 'Stretching Contrast Limits,' extends the contrast of the RGB image by calculating and applying the upper and lower limits of each band. LodgeNet, which relies on the state-of-the-art YOLOv8 deep learning algorithm, could detect and classify wheat lodging severity levels ranging from no lodging (Class 1) to severe lodging (Class 9). The results show the mean Average Precision (mAP) of 0.952% @0.5 and 0.641% @0.50-0.95 in classifying wheat lodging severity levels. LodgeNet promises an efficient and automated high-throughput solution for real-time crop monitoring of wheat lodging severity levels in the field.

Exploring the impact of PVC and PVA microplastics on zebrafish tissue using multi-spectral imaging, Optical Coherence Tomography (OCT) and biospeckle OCT (bOCT).

Plastics are widely used in industry and households, but improper disposal has caused their accumulation in aquatic systems worldwide. As a result, mechanical and photochemical processes break down these plastics into microplastics or nano plastics, posing a severe threat to marine organisms and humans as they enter the food chain. This study investigates the effect of Polyvinyl chloride (PVC) and Polyvinyl alcohol (PVA) microplastics in zebrafish by using multi-spectral imaging (MSI), Optical Coherence Tomography (OCT), and Biospeckle OCT (bOCT). These techniques allow for long-term studies in the fish without invasive procedures in real-time. Zebrafish were exposed to Nile red labeled PVC and PVA for 21 days with 500mg/L concentration. Image acquisition and analysis were performed every five days till the end of the study. MSI images revealed deposition of microplastics in the gills region of the fish; some diffused deposition was seen throughout the body in the PVA group towards the end of the experiment. The effect of these MPs on the structure of the gills and their exact location was determined by capturing OCT images. bOCT was used to determine the average speckle contrast for all the OCT images to determine the change in biological activity within the gills region. An increase in bioscpeckle contrast was observed for the MPs treated groups compared to the control group. PVC appeared to cause a more considerable rise in activity compared to PVA. The results indicated that the MPs exert stress on the gills and increase activity within the gills, possibly due to the blockage of the gills and disruption of the water filtration process, which could be monitored non-invasively only by using bOCT. Overall, our study demonstrates the usefulness of non-invasive, robust techniques like MSI, bOCT, and biospeckle for long-term zebrafish studies and real-time analyses.

Assessment of Landsat-8 and Sentinel-2 Water Indices: A Case Study in the Southwest of the Buenos Aires Province (Argentina).

The accuracy assessment of three different Normalized Difference Water indices (NDWIs) was performed in La Salada, a typical lake in the Pampean region. Data were gathered during April 2019, a period in which floods occurred in a large area in the Southwest of the Buenos Aires Province (Argentina). The accuracy of the estimations using spaceborne medium-resolution multi-spectral imaging and the reliability of three NDWIs to highlight shallow water features in satellite images were evaluated using a high-resolution airbone imagery as ground truth. We show that these indices computed using Landsat-8 and Sentinel-2 imagery are only loosely correlated to the actual flooded area in shallow waters. Indeed, NDWI values vary significantly depending on the satellite mission used and the type of index computed.

A modulation method that can improve the performance of LED multi-spectral imaging.

In the LED multi-spectral imaging (LEDMSI) system, modulation by using the square wave frequency division with frequency ratio of 2 can improve the image quality and acquisition speed, but it will occupy a wide frequency band. Moreover, since there is a simultaneous change in the state of multiple signals when this method is used, it will lead to serious ringing phenomenon or insufficient slew rate and affect the quality of multi-spectral images. To solve the above problems, this paper proposes a modulation method for LEDMSI system, which uses n same frequency square wave signals with different phases as the carrier signals. Comparing the multi-spectral images modulated by the proposed method with the multi-spectral images modulated by the traditional method, experimental results show that the quality of the image modulated by the proposed method is higher, which indicates that the proposed method is of great significance to improve the performance of LEDMSI system.

Open-source mobile multispectral imaging system and its applications in biological sample sensing.

Visible-near-infrared spectroscopy data can be utilized as an important quantitative indicator of biomolecular quantitative analysis. When acquiring spectral information, hyperspectral/multispectral imaging systems can obtain the spatial information of the object of interest. This allows the complete spatial-spectral information of the object of interest to be acquired and the spatial distribution of biomolecules to be analyzed. In this study, we present an open-source mobile multispectral imaging system, test the influence of the utilization of LEDs on the multispectral image, and design image-processing algorithms to correct this influence. Todemonstrate the effectivenessofthesystem, the system is applied to meat freshness analysis, small-animal tumor in-vivo imaging, and chlorophyll spatial distribution imaging. The experimental results verify that our system has stable performance and is compatible with a wide range of spectral imaging applications.

Multi-spectral imaging in adult-onset foveomacular vitelliform dystrophy: Report of two cases.

Purpose: To describe the characteristic findings of non-invasive multi-spectral imaging (MSI) for adult-onset foveomacular vitelliform dystrophy (AFVD). Observations: On examination of MSI, the characteristic performances of AFVD include the nodule-like high-reflecting lesions, the line-like low-reflecting lesions in the high-reflecting lesion, and the scattered high-reflecting and low-reflecting lesions around the nodule-like lesion. MSI has an advantage over color fundus photography (CFP) and fundus autofluorescence (FAF) in finding tiny lesions, which corresponded to drusenoid structures on optical coherence tomography (OCT). MSI showed different characteristics at different stages of AFVD, which may be instructive to the pathogenesis and progression of AFVD. Conclusions and Importance: MSI is a promising diagnostic and follow-up tool that will provide additional information in fundus imaging for AFVD, and the changes on MSI is partially instructive to the pathogenesis and progression of AFVD.

Non-destructive assessment of quality parameters of white button mushrooms (Agaricus bisporus) using image processing techniques.

Considering that appearance of white button mushroom (WBM) as the trigger for registering its quality, this study was aimed at analyzing the visual cues by the application of image processing tools. While L-a-b colour space and skewness was used for estimating chromatic and morphological characteristics; onset of discolouration of WBM was predicted by hyperspectral image analysis. Undamaged (UD) and damaged (D) mushrooms were stored under refrigerated conditions (3-5 °C and 90% Rh). RGB and hyperspectral images were acquired on alternate storage days 1, 3, 5, 7 and 9. Weight loss, texture and moisture content of stored mushrooms were also recorded during the storage period. Colour changes in stored UD and D were found to be in b (21.55) and a (2399) value, respectively. Browning index in D was 83-212% higher than UD mushrooms across the storage period. Weight and firmness losses in D were higher by 65.9 and 31.4%, respectively than UD. Morphological characteristic in terms of aspect ratio and roundness were not found to vary significantly over the storage period for both UD and D mushrooms. Chemometrics revealed that multiplicative scatter correction was the best pre-processing tool and that onset on discolouration is conspicuous in the spectral region of 520-800 nm. k-NN fared better than PLS-DA for correct classification (100%) of UD and D mushrooms.

Comprehensive imaging of microcirculatory changes in the foot during endovascular intervention - A technical feasibility study.

Chronic limb-threatening ischemia (CLTI) has a major impact on patients' lives and is associated with a heavy health care burden with high morbidity and mortality. Treatment by endovascular intervention is mostly based on macrocirculatory information from angiography and does not consider the microcirculation. Despite successful endovascular intervention according to angiographic criteria, a proportion of patients fail to heal ischemic lesions. This might be due to impaired microvascular perfusion and variations in the supply to different angiosomes. Non-invasive optical techniques for microcirculatory perfusion and oxygen saturation imaging have the potential to provide the interventionist with additional information in real-time, supporting clinical decisions during the intervention. This study presents a novel multimodal imaging system, based on multi-exposure laser speckle contrast imaging and multi-spectral imaging, for continuous use during endovascular intervention. The results during intervention display spatiotemporal changes in the microcirculation compatible with expected physiological reactions during balloon dilation, with initially induced ischemia followed by a restored perfusion, and local administration of a vasodilator inducing hyperemia. We also present perioperative and postoperative follow-up measurements with a pulsatile microcirculation perfusion. Finally, cases of spatial heterogeneity in the oxygen saturation and perfusion are discussed. In conclusion, this technical feasibility study shows the potential of the methodology to characterize changes in microcirculation before, during, and after endovascular intervention.

Beef Cut Classification Using Multispectral Imaging and Machine Learning Method.

Classification of beef cuts is important for the food industry and authentication purposes. Traditional analytical methods are time constraints and incompatible with the modern food industry. Taking advantage of its rapidness and being nondestructive, multispectral imaging (MSI) has been widely applied to obtain a precise characterization of food and agriculture products. This study aims at developing a beef cut classification model using MSI and machine learning classifiers. Beef samples are imaged with a snapshot multi-spectroscopic camera within a range of 500-800 nm. In order to find a more accurate classification model, single- and multiple-modality feature sets are used to develop an accurate classification model with different machine learning-based classifiers, namely, linear discriminant analysis (LDA), support vector machine (SVM), and random forest (RF) algorithms. The results demonstrate that the optimized LDA classifier achieved a prediction accuracy of over 90% with multiple modality feature fusion. By combining machine learning and feature fusion, the other classification models also achieved a satisfying accuracy. Furthermore, this study demonstrates the potential of machine learning and feature fusion method for meat classification by using multiple spectral imaging in future agricultural applications.

Fast Detection of Olive Trees Affected by Xylella Fastidiosa from UAVs Using Multispectral Imaging.

Xylella fastidiosa (Xf) is a well-known bacterial plant pathogen mainly transmitted by vector insects and is associated with serious diseases affecting a wide variety of plants, both wild and cultivated; it is known that over 350 plant species are prone to Xf attack. In olive trees, it causes olive quick decline syndrome (OQDS), which is currently a serious threat to the survival of hundreds of thousands of olive trees in the south of Italy and in other countries in the European Union. Controls and countermeasures are in place to limit the further spreading of the bacterium, but it is a tough war to fight mainly due to the invasiveness of the actions that can be taken against it. The most effective weapons against the spread of Xf infection in olive trees are the detection of its presence as early as possible and attacks to the development of its vector insects. In this paper, image processing of high-resolution visible and multispectral images acquired by a purposely equipped multirotor unmanned aerial vehicle (UAV) is proposed for fast detection of Xf symptoms in olive trees. Acquired images were processed using a new segmentation algorithm to recognize trees which were subsequently classified using linear discriminant analysis. Preliminary experimental results obtained by flying over olive groves in selected sites in the south of Italy are presented, demonstrating a mean Sørensen-Dice similarity coefficient of about 70% for segmentation, and 98% sensitivity and 93% precision for the classification of affected trees. The high similarity coefficient indicated that the segmentation algorithm was successful at isolating the regions of interest containing trees, while the high sensitivity and precision showed that OQDS can be detected with a low relative number of both false positives and false negatives.

Plenoptic Face Presentation Attack Detection.

The vulnerability of current face recognition systems to presentation attacks significantly limits their application in biometrics. Herein, we present a passive presentation attack detection method based on a complete plenoptic imaging system which can derive the complete plenoptic function of light rays using a single detector. Moreover, we constructed a multi-dimensional face database with 50 subjects and seven different types of presentation attacks. We experimentally demonstrated that our approach outperforms the state-of-the-art methods on all types of presentation attacks.

Multi-spectral laser speckle contrast images using a wavelength-swept laser.

A multi-spectral laser speckle contrast imaging (MS-LSCI) system is proposed using only a single wavelength-swept laser, which provides both highly coherent and multi-spectral outputs to simultaneously generate laser speckle contrast images and multi-spectral images, respectively. Using a laser light swept from 770 to 821 nm at a repetition rate of 5 Hz and a CCD camera of 335 fps, 67 multi-spectral frame images are acquired in 0.76 nm wavebands over 51 nm spectral range. The spectral sub-windowing method of single wavelength-swept laser source is used to solve the lack of spectral information from a few individual light sources, which is a limitation of conventional MS-LSCI systems. In addition to the speckle flow index from the LSCI frames, the multi-spectrally encoded images can generate additional images of spectral absorbance. To further examine the performance of the MS-LSCI system, an in vivo cuff-induced ischemia experiment was conducted to show the real-time imaging of hemodynamic and blood oxygen saturation changes simultaneously over the entire 2.5 cm × 4.5 cm field of view.

Design and Fabrication of an Artificial Compound Eye for Multi-Spectral Imaging.

The artificial compound eye (ACE) structure is a new type of miniaturized, lightweight and intelligent imaging system. This paper has proposed to design a multi-spectral ACE structure to enable the structure to achieve multi-spectral information on the basis of imaging. The sub-eyes in the compound eye structure have been designed as diffractive beam splitting lenses with the same focal length of 20 mm, but with the different designed center wavelengths of 650 nm, 532 nm, and 445 nm, respectively. The proximity exposure lithography and reactive ion etching process were used to prepare the designed multi-spectral ACE structure, and the spectral splitting and multi-spectral imaging experiments were carried out to verify the multi-spectral imaging function of the structure without axial movement. Furthermore, the structure can be designed according to actual requirements, which can be applied to covert reconnaissance, camouflage identification, gas leakage or other fields.

A Novel Illumination Compensation Technique for Multi-Spectral Imaging in NDVI Detection.

To overcome the dependence on sunlight of multi-spectral cameras, an active light source multi-spectral imaging system was designed and a preliminary experimental study was conducted at night without solar interference. The system includes an active light source and a multi-spectral camera. The active light source consists of four integrated LED (Light Emitting Diode) arrays and adjustable constant current power supplies. The red LED arrays and the near-infrared LED arrays are each driven by an independently adjustable constant current power supply. The center wavelengths of the light source are 668 nm and 840 nm, which are consistent with that of filter lens of the Rededge-M multi-spectral camera. This paper shows that the radiation intensity measured is proportional to the drive current and is inversely proportional to the radiation distance, which is in accordance with the inverse square law of light. Taking the inverse square law of light into account, a radiation attenuation model was established based on the principle of image system and spatial geometry theory. After a verification test of the radiation attenuation model, it can be concluded that the average error between the radiation intensity obtained using this model and the actual measured value using a spectrometer is less than 0.0003 w/m2. In addition, the fitting curve of the multi-spectral image grayscale digital number (DN) and reflected radiation intensity at the 668 nm (Red light) is y = -3484230x2 + 721083x + 5558, with a determination coefficient of R2 = 0.998. The fitting curve with the 840 nm (near-infrared light) is y = 491469.88x + 3204, with a determination coefficient of R2 = 0.995, so the reflected radiation intensity on the plant canopy can be calculated according to the grayscale DN. Finally, the reflectance of red light and near-infrared light can be calculated, as well as the Normalized Difference Vegetation Index (NDVI) index. Based on the above model, four plants were placed at 2.85 m away from the active light source multi-spectral imaging system for testing. Meanwhile, NDVI index of each plant was measured by a Greenseeker hand-held crop sensor. The results show that the data from the two systems were linearly related and correlated with a coefficient of 0.995, indicating that the system in this article can effectively detect the vegetation NDVI index. If we want to use this technology for remote sensing in UAV, the radiation intensity attenuation and working distance of the light source are issues that need to be considered carefully.

A rapid monitoring of NDVI across the wheat growth cycle for grain yield prediction using a multi-spectral UAV platform.

Wheat improvement programs require rapid assessment of large numbers of individual plots across multiple environments. Vegetation indices (VIs) that are mainly associated with yield and yield-related physiological traits, and rapid evaluation of canopy normalized difference vegetation index (NDVI) can assist in-season selection. Multi-spectral imagery using unmanned aerial vehicles (UAV) can readily assess the VIs traits at various crop growth stages. Thirty-two wheat cultivars and breeding lines grown in limited irrigation and full irrigation treatments were investigated to monitor NDVI across the growth cycle using a Sequoia sensor mounted on a UAV. Significant correlations ranging from R2 = 0.38 to 0.90 were observed between NDVI detected from UAV and Greenseeker (GS) during stem elongation (SE) to late grain gilling (LGF) across the treatments. UAV-NDVI also had high heritabilities at SE (h2 = 0.91), flowering (F)(h2 = 0.95), EGF (h2 = 0.79) and mid grain filling (MGF) (h2 = 0.71) under the full irrigation treatment, and at booting (B) (h2 = 0.89), EGF (h2 = 0.75) in the limited irrigation treatment. UAV-NDVI explained significant variation in grain yield (GY) at EGF (R2 = 0.86), MGF (R2 = 0.83) and LGF (R2 = 0.89) stages, and results were consistent with GS-NDVI. Higher correlations between UAV-NDVI and GY were observed under full irrigation at three different grain-filling stages (R2 = 0.40, 0.49 and 0.45) than the limited irrigation treatment (R2 = 0.08, 0.12 and 0.14) and GY was calculated to be 24.4% lower under limited irrigation conditions. Pearson correlations between UAV-NDVI and GY were also low ranging from r = 0.29 to 0.37 during grain-filling under limited irrigation but higher than GS-NDVI data. A similar pattern was observed for normalized difference red-edge (NDRE) and normalized green red difference index (NGRDI) when correlated with GY. Fresh biomass estimated at late flowering stage had significant correlations of r = 0.30 to 0.51 with UAV-NDVI at EGF. Some genotypes Nongda 211, Nongda 5181, Zhongmai 175 and Zhongmai 12 were identified as high yielding genotypes using NDVI during grain-filling. In conclusion, a multispectral sensor mounted on a UAV is a reliable high-throughput platform for NDVI measurement to predict biomass and GY and grain-filling stage seems the best period for selection.