This PDF file contains the front matter associated with SPIE Proceedings Volume 12545, including the Title Page, Copyright information, Table of Contents, and Conference Committee information.

Sharjah-Sat-1 is the first CubeSat to be designed and integrated at the Sharjah Academy for Astronomy, Space Sciences and Technology (SAASST), a research institute under the University of Sharjah (UoS) in the United Arab Emirates, with an active collaboration with Istanbul Technical University and Sabanci University in Turkey. The mission is due to launch in December 2022. Sharjah-Sat-1 hosts a primary payload of an improved X-Ray Detector (iXRD). The iXRD utilizes a CdZnTe crystal as an active detector to detect and measure bright and hard X-Ray sources and a tungsten collimator. The instrument’s detection range is 20-200 KeV at a spectral resolution of 6 Kev at 60 KeV [1]. The detector will be able to measure the flux of ionizing x-ray around the south Atlantic anomaly, the data of which will be shared to provide space situational awareness for other satellite operators to perform any preventative maneuvers to protect their space assets. This paper will discuss how the improved X-Ray Detector (iXRD) on-board the Sharjah-Sat-1 CubeSat can be utilized to provide space situational awareness.

Rice is a major world food staple and there is increasing interest in producing rice grains with pigmented bran colors, black/purple or red, which are rich in antioxidants, providing human health benefits. Identification of bran color to perform breeding selections requires the removal of the outer hull, which is a destructive process. Being able to detect bran color without dehulling would have advantages in breeding as well as in quality control of seed rice production for both brown and colored bran varieties. In this study, we explored single-kernel NIR spectroscopy and NIR hyperspectral imaging for rapid and non-destructive prediction of rice bran color. Color (L*, a*, and b*) values of dehulled rice samples were measured and rough rice samples were scanned using SKNIR and NIR hyperspectral imaging. The prediction results showed that SKNIR and hyperspectral imaging can be potentially used for efficient sorting of rough rice according to bran color.

A multimodal sensing system was developed for automated and intelligent food safety inspection. The system uses two pairs of lasers and spectrometers at 785 and 1064 nm to realize dual-band Raman measurement. Automated sampling can be conducted using a XY moving stage for solid, powder, and liquid samples in customized well plates or randomly scattered in standard Petri dishes (e.g., bacterial colonies). Three LED lights (white backlight, UV ring light, and white ring light) and two color cameras are used for machine vision measurements of samples in the Petri dishes (e.g., transmission, fluorescence, and color). Real-time image processing and motion control techniques are used to fulfill automated sample counting, positioning, sampling, and synchronization functions. System software was developed with integrated AI functions able to identify and label interesting targets instantly. The system capability was demonstrated by an example application for rapid identification of five common foodborne bacteria. Using a machine learning model based on a linear support vector machine, a classification accuracy of 98.6% was achieved using Raman spectra collected from bacterial colonies grown on nutrient nonselective agar in Petri dishes. The system is compact and portable (30×45×35 cm³) that can be used for biological and chemical food safety inspection in regulatory and industrial applications.

In the almond industry, the presence of bitter almonds in processed batches is a common problem that causes not only unpleasant flavors but also problems in the product commercialization. This research group has previously demonstrated the potential of Near Infrared Spectroscopy (NIRS) to detect adulterated almond batches; however, since NIRS provides an average spectrum of each batch, it does not enable to identify each individual bitter almond. Hyperspectral Imaging (HSI), which integrates both the spectral and spatial dimensions, enables to know the spatial distribution of the different physico-chemical characteristics, favoring the individual identification of the different compounds in the sample. The aim of this study was to evaluate the feasibility of using a HSI system for the identification of bitter almonds in sweet almond batches. Samples were analyzed using a HSI camera working in the spectral range 946.6–1648.0 nm and Partial Least Squares Discriminant Analysis (PLS-DA) was applied. A classification success over the 99% was obtained in cross-validation and the pixel-by-pixel validation identified correctly between the 61 – 85% of the adulterations. The results confirm that HSI can be considered a promising approach for the classification of almonds by bitterness, allowing the identification of each single bitter almond present in the batch.

Marbling is one of the most important determinants for beef quality and assessed in terms of the abundance and spatial distribution of visible fat flecks in the longissimus dorsi (LD) muscle. Visual appraisal by trained professionals is the currently prevailing practice in beef marbling assessment, but it suffers from the drawbacks of human subjective, laborintensive, and time-consuming. Computer vision technology offers a promising alternative for objective and automated beef quality assessment. Recently, imaging technology under structured illumination has emerged for the detection of meat quality characteristics as opposed to existing imaging modalities using uniform illumination. By modulating light at certain spatial frequencies, structured illumination reflectance imaging has the potential to resolve subtle texture features of meat surface/subsurface for enhanced quality assessment. This study represents the proof-of-concept evaluation of the applicability of an inhouse-assembled structured illumination imaging system, combined with deep learning, for beef marbling assessment. Beef samples of varying marbling degrees were imaged under sinusoidal illumination at spatial frequencies of 0.05-0.40 cycles/mm. The acquired images were demodulated into direct component (DC) and amplitude component (AC) images at each spatial frequency. A deep learning segmentation model, SegFormer, was built using DC and AC (0.05-0.40 cycles/mm) images for segmenting the LD muscle. Texture features were extracted by a pretrained ResNext model from the LD muscle segments, and then used for building discriminative models to classify samples of three marbling categories. The DC images yielded the overall classification accuracy of 76.32%, while the AC images resulted in improved accuracies of 76.84%-81.05%, which the best accuracy attained at the spatial frequency of 0.40 cycles/mm. This study shows the effectiveness of imaging under sinusoidal illumination in place of uniform illumination for enhanced assessment of beef marbling.

Protecting elders in long-term care facilities (LTCFs) from foodborne illnesses, such as norovirus, listeria, salmonella, and E. coli, is critical. Sanitation inspection is an ongoing concern for LTCF kitchens and dining facilities and staff who handle and serve food. LTCFs must prevent food contamination but must also deal with the potential spread of infection among workers and customers. By 2050 the number of Americans needing LTCFs is expected to double. The Centers for Disease Control and Prevention (CDC) reports that 1 to 3 million serious infections occur annually in nursing homes and assisted living. LTCF sanitization can benefit from standardized tools such as checklists and frequent staff education, including specific product use training. Visual inspection is the most common evaluation method for cleanliness after cleaning but is non-objective and less accurate. Swab-based adenosine triphosphate (ATP) bioluminescence assays are objective for evaluating the quality of cleaning in LTCFs. While more accurate than visual assessment, it requires additional swab and analysis time. We present a fast and easy-to-use handheld fluorescence imaging system for infection prevention in LTCFs. It detects invisible contamination, provides immediate UVC deactivation of potential threats (i.e., bacteria, viruses), and documentation for traceable evidence of cleanliness. We have developed an algorithm to detect organic residue contamination found in images of high-touch surfaces. We provide fluorescence imaging optimization of camera parameters to improve the machine-learning results of different surfaces in LTCFs that were measured, analyzed, and recorded. This information can improve cleaning procedures and educate and train staff.

Escherichia coli (E. coli) bacteria when transmitted to humans by infected fruits like citrus can cause serious illnesses like bloody diarrhea and kidney disease (Hemolytic Uremic Syndrome). Therefore, inspection for the presence of E.coli colonies on fruits and vegetables would be beneficial for food safety if an appropriate sensor and detection approach were available. In this paper, we evaluate the efficacy of SafetySpect’s Contamination, Sanitization Inspection, and Disinfection (CSI-D+) system, which consists of an UV camera, an RGB camera, and illumination at two fluorescence excitation wavelengths, ultraviolet C (UVC) at 275 nm and violet at 405 nm, respectively. For this study, bacterial population at four different concentration levels were inoculated on black rubber slides. The black rubber slides provided a fluorescence-free background for benchmark tests on E.coli-containing droplets, since even fruit peels have a subdued fluorescence response which appears as low intensity noise in the background of the UV images. To increase the number of image datasets to avoid overfitting issue of deep leaning and machine learning models, synthetic datasets were generated using StyleGAN2- ADA generative network. As a first approach, the fluorescence images were denoised using discrete wavelet transform (DWT) and reconstructed denoised images were used to train StyleGAN2-ADA for generating larger datasets which were later used by VGG19 with SoftMax classifier. In the second approach, VGG19 was used to extract features and then Linear and RBF SVM were used for classification. It was found that using Symlets family of DWT denoised images at an average PSNR value of 25.59 dB. VGG19 with SoftMax was able to classify the images at overall accuracy of 84% without synthetic dataset and at 94% by using augmented dataset generated by StyleGAN2-ADA. It was also found that using RBF SVM increased the accuracy by 2% to 96% and linear SVM by 3% to 97%. The findings reported here could be used to detect presence of E. coli bacterial population on citrus peels for taking the necessary actions for decontamination.

The consumption of mycotoxins generated by fungi can have severe effects on the health of both humans and animals. These toxins can exist at dangerous levels in food products made from crops that have been infected with mycotoxinproducing fungi. Numerous methods have been developed for detecting mycotoxins in order to divert contaminated commodities from the food supply, but only allow for reactive, not preventive approaches. Furthermore, under favorable conditions toxin-producing fungi can continue to produce mycotoxins during storage and throughout the crop processing stages. By identifying mycotoxin-producing fungal species on crops or commodities, remediation such as fungicide application can be carried out, preventing the spread of infection and potential contamination of healthy crops, reducing waste of resources and ultimately improving food safety. Loop-mediated isothermal amplification (LAMP) has advantages for portable DNA detection due to its isothermal nature, resistance to matrix inhibitors, and the possibility of a long shelflife when reagents are dried onto a matrix. The developed microfluidic device allows for the homogenized wheat sample input after DNA extraction. The microfluidic device functions as a disposable cassette and can be heated by an independent, portable, isothermal heating device. The LAMP assay is combined with calcein for fluorescence detection. In this experiment, Fusarium graminearum, a trichothecene mycotoxin producer, was used as a proof-of-concept for the device with a LAMP assay targeting the gaoA gene, which codes for the enzyme galactose oxidase (GO), a unique enzyme produced by only a few other fungal species. The presence of Fusarium graminearum was detected in contaminated wheat samples utilizing the described methods, indicating the potential detection of mycotoxin-producing fungi. In the future, the device will be expanded to test for multiple mycotoxin-producing genes.

Recently, the use of a Quartz Crystal Microbalance (QCM) as a biosensor for detecting foodborne pathogens by observing changes in resonant frequency has gained popularity. However, conventional detection methods are time-consuming and require expensive equipment and trained personnel. The current trend is toward detection approaches that are quick, portable, and easy to use. In order to address this need, a dual-modality QCM system combining a smartphone, an in-situ fluorescence imaging subsystem, and a flow injection component has been proposed. This system enables a smartphone to receive real-time frequency data via Bluetooth, while a camera detects the presence of bacteria on the quartz crystal surface using a fluorescence-tagged antibody. The fluorescence imaging subsystem utilizes a camera to capture the bacteria fluorescence signal, while the flow injection subsystem employs a mini peristaltic pump and controller to introduce biochemical solutions, antibodies, and bacteria. All components are contained in a 3D cartridge that is portable. FITC images were captured with 5 MHz quartz crystals when the prototype system was tested. The developed QCM biosensor has the potential to become a portable bacteria detection approach that outperforms existing techniques.

Due to the increasing complexity of the food supply chain, the likelihood of food adulteration or contamination is a significant problem for food safety. Agricultural products may be chemically, physically, or biologically manipulated at numerous points along the supply chain. To address this issue and improve food safety, it is necessary to implement innovative measurement technologies for biohazard detection. Currently, accessible food analysis techniques include vibrational spectroscopy and mass spectrometry. However, optical techniques such as laser-induced breakdown spectroscopy (LIBS) and Raman systems are gaining popularity due to their real-time analysis capabilities and minimal requirements for sample preparation. In this study, we combined LIBS and Raman detection to analyze the elemental and molecular composition of various food matrices for the purpose of detecting food contamination in real time. We examined typical generic herbicides containing glyphosate, such as Roundup. The samples were spiked by spraying the chemical compound on the surface of fruits. The results revealed that it is possible to assess complex food matrices polluted with widespread organic contaminants by combining optical spectroscopies very rapidly.

S. Typhi was a common foodborne pathogen that often enters the human body through food and caused illness. In order to achieve the rapid detection of S. Typhi, surface enhanced Raman spectroscopy (SERS) had been studied by more and more people. To prepare an efficient, stable, and simple signal enhancer, the signal enhancement effects of Au NPs, Ag NPs, and Au @ Ag NPs on S. Typhi were compared. Because of the good enhancement effect and stability, Au @ Ag NPs was used as the research object. The influence of SERS substrates with the same gold core size and different silver shell thickness on bacterial signal enhancement was investigated. Au @ Ag NPs with different thickness silver layers were prepared by changing the concentration of silver nitrate, and colloids were characterized by UV spectroscopy. After de-noising, smoothing and de-baselining of SERS spectra, the characteristic peak intensity information was used to analyze the detection limit and detection stability of S. Typhi. The experimental results showed that when the silver layer thickness was prepared by 4 mM silver nitrate, Au @ Ag NPs had the best amplification effect on S. Typhi Raman signal, which could achieve a detection limit of 104 CFU/mL, and the coefficient of variation of SERS characteristic peak was less than 7%. In addition, the quantitative prediction model of S. Typhi with 10⁵ CFU/mL - 10⁸ CFU/mL was established by MLR, which achieved a correlation coefficient (Rc) of 0.99 and a root mean square error (RMSEC) of 10^0.0364 CFU/mL.

Mississippi and Alabama are the top two states producing and processing catfish in the United States, with the annual production of $382 million in 2022. The catfish industry supplies protein-rich catfish products to the U.S. market and contributes considerably to the development of the local economy. However, the traditional catfish processing heavily relies on human labors leading to a high demand of workforce in the processing facilities. De-heading, gutting, portioning, filleting, skinning, and trimming are the main steps of the catfish processing, which normally require blade-based cutting device (e.g., metal blades) to handle. The blade-based manual catfish processing might lead to product contamination, considerable fish meat waste, and low yield of catfish fillet depending on the workers’ skill levels. Furthermore, operating the cutting devices may expose the human labors to undesired work accidents. Therefore, automated catfish cutting process appears to be an alternative and promising solution with minimal involvement of human labors. To further enable, assist, and automate the catfish cutting technique in near real-time, this study presents a novel computer vision-based sensing system for segmenting the catfish into different target parts using deep learning and semantic segmentation. In this study, 396 raw and augmented catfish images were used to train, validate, and test five state-of-the-art deep learning semantic segmentation models, including BEiTV1, SegFormer-B0, SegFormer-B5, ViT-Adapter and PSPNet. Five classes were pre-defined for the segmentation, which could effectively guide the cutting system to locate the target, including the head, body, fins, tail of the catfish, and the image background. Overall, BEiTV1 demonstrated the poorest performance with 77.3% of mIoU (mean intersection-over-union) and 86.7% of MPA (mean pixel accuracy) among all tested models using the test data set, while SegFormer-B5 outperformed all others with 89.2% of mIoU and 94.6% of MPA on the catfish images. The inference speed for SegFormer-B5 was 0.278 sec per image at the resolution of 640x640. The proposed deep learning-based sensing system is expected to be a reliable tool for automating the catfish cutting process.

In brewing business more and more consistent quality, coupled with innovation and production cost reduction, is required. One of the most important parameters to be monitored during fermentation is density, being directly correlated with the alcohol and sugar content. The traditional method used to measure beer density requires the utilization of a hydrometer. In this study, a simple and fast approach to assess sugar content in hopped wort of artisanal beer, based on the utilization of a portable spectroscopic device working in the Short-Wave InfraRed (SWIR) region (1000-2400 nm), is adopted in order to be utilized both off- and on-line. The proposed approach, faster than the traditional hydrometric method, will allow to realize a better control of the process, reducing production cost and increasing, at the same time, product quality. Starting from the collected spectra, acquired in transreflectance mode and the reference density data, Partial Least Square (PLS) regression models able to predict the density of hopped wort were developed (i.e., Rp 2=0.98 and RMSEP = 0.3 °P; Rp ²=0.97 and RMSEP=0.5 °P, after VIP scores selection). PLS regression was optimized by pre– processing algorithms using an ad hoc code to achieve optimal performances and the computational time was reduced performing a wavelength reduction by VIP scores method.

Escherichia coli and Salmonella enterica are significant causes of gastrointestinal disease globally. The Contamination Sanitization Inspection and Disinfection (CSI-D) device is a handheld fluorescence-based imaging system that disinfects food contact surfaces using ultraviolet-C (UVC) illumination. The goal of this study was to determine the optimal parameters for disinfection of E. coli and S. enterica using the CSI-D system. E. coli and S. enterica Enteritidis, Newport, Typhimurium, and Javiana were grown on selective media, followed by transfer to Luria Bertani broth. After overnight incubation, the cultures were diluted and spread-plated on L-agar. The plates were exposed to high-intensity (10 mW/cm2) or low-intensity (5 mW/cm2) UVC for 1 s, 3 s, or 5 s. Exposed and control plates were incubated at room temperature for 2-3 h, then overnight at 37°C. The resulting colonies were counted and compared to control plates. Three trials were conducted on separate days. The average of the trials showed that exposure times of 3-5 s at either intensity resulted in effective and consistent destruction of E. coli and S. enterica. The minimum reduction at 3 s exposure for both intensities was 96%, with a maximum of 100%. The 1 s exposure time showed inconsistent results, with a 0-61.5% survival rate. The results of this study show that exposure to UVC for at least 3 s is required to achieve consistent disinfection of 96-100% for generic E. coli and S. enterica.

SafetySpect previously reported the development of a handheld fluorescence imaging system for contamination detection in food facilities using an excitation wavelength of 405 nm. Here we report a dual-excitation fluorescence imaging system including 365 nm and 405 nm LED illumination and a dual camera system for multi-band fluorescence emission image detection. We have measured fluorescence excitation and emission spectra for multiple food residues at the USDA/ARS research laboratory using a Spex spectrofluorometer (JY Inc., Edison, New Jersey). By analyzing emission spectra corresponding to 365 nm and 405 nm excitation wavelengths, we have optimized multi-bandpass optical filters used in our camera systems. We collected image data from multiple kitchen facilities to validate fluorescence imaging capabilities for contamination detection in real-world conditions. Food residue samples evaluated in a laboratory setting included multiple samples of fish, scallops, shrimp, chicken, pork, beef, lamb, fruits, nuts, vegetable oils, condiments, and starches. System performance measurements of the imaging system that were evaluated include characteriz ation of the field of view, working distance, image resolution, image registration, exposure timing accuracy, camera/filter system wavelength response, and dynamic range. System performance measurements of the illumination system that were evaluated includ e wavelength characteristics, exposure control accuracy, camera synchronization accuracy, optical power, linearity of optical power with exposure setting, the field of view, and field uniformity.

Protecting patients receiving care in healthcare facilities from contracting illnesses such as C. difficile, S. aureus, and Acinetobacter is critical. The bed or room incoming patients are pla ced in dramatically affects their chances of contracting an illness from a previous patient. The Centers for Disease Control and Prevention (CDC) reports that 1 in 31 hospital patients have at least one healthcare-associated infection (HAI). Surveillance for HAI is a priority given that the patients receiving healthcare often have compromised immune status. Currently , no real-time tool to monitor cleanliness efficacy provides information for immediate mitigation in large-scale institutional environments. According to the CDC, the transmission of many healthcare-acquired pathogens is related to the contamination of near-patient surfaces and equipment. Therefore, hospitals are encouraged to optimize and improve high-touch surface cleaning at the time of discharge or transfer of pa tients. We have developed a fast and easy-to-use scanner that objectively assesses cleanliness and cleaning product efficacy in healthcare facilities. The scanner detects invisible contamination, provides UVC disinfection of any contamination identified that may harbor bacteria and viruses, and an audit trail of image data for evidence of cleanliness. In addition, we have developed an image segmentation algorithm that provides live identification and labeling of organic residue contamination in video images of high-touch surfaces. Finally, we present fluorescence imaging results of different surfaces in healthcare that were measured, analyzed, and recorded. This information can be used to improve cleaning procedures and for staff education and training.

This research paper presents an AI-based insect detection system that uses an affordable and power-saving selfcontained computer - the Jetson Nano, a manual focus camera, and a trained Convolutional Neural Network (CNN). The system addresses the need for real-time monitoring and detection of insect pests in grain storage and food facilities, which is crucial for effective insect control and decision-making. The camera-based monitoring system employs CNN to detect and identify small-scale stored grain insect pests. The Jetson Nano processes insect images captured by the camera using the trained machine learning model. The system's effectiveness is evaluated by computing F1 scores, and the accuracy is analyzed under varying illumination settings, including white LED light, yellow LED light, and the absence of any light source. Taking adult warehouse beetles (Trogoderma variabile) and cigarette beetles (Lasioderma serricorne (F.)) as test cases, the system was found to accurately detect the presence and type of insects, making it an affordable and efficient solution for identifying and monitoring insect infestations in stored product facilities. This automated insect detection system can reduce pest control costs, save producers time and energy, and maintain product quality. The proposed system offers a practical solution for automated insect detection in grain storage and food facilities. The low-cost and low-power Jetson Nano makes the system affordable and accessible for system developers and ultimately for a wide range of producers. The system's ability to detect and identify insect pests in real time enables quick decision-making and effective pest control management. The results demonstrate that the proposed system is a promising approach for automated insect detection and monitoring in stored product facilities.