Search Results (41)

Real-time quality monitoring using molten pool images is a critical focus in researching high-quality, intelligent automated welding. To address interference problems in molten pool images under complex welding scenarios (e.g., reflected laser spots from spatter misclassified as porosity defects) and the limited interpretability of deep learning models, this paper proposes a multi-granularity spatiotemporal representation learning algorithm based on the hybrid enhancement of handcrafted and deep learning features. A MobileNetV2 backbone network integrated with a Temporal Shift Module (TSM) is designed to progressively capture the short-term dynamic features of the molten pool and integrate temporal information across both low-level and high-level features. A multi-granularity attention-based feature aggregation module is developed to select key interference-free frames using cross-frame attention, generate multi-granularity features via grouped pooling, and apply the Convolutional Block Attention Module (CBAM) at each granularity level. Finally, these multi-granularity spatiotemporal features are adaptively fused. Meanwhile, an independent branch utilizes the Histogram of Oriented Gradient (HOG) and Scale-Invariant Feature Transform (SIFT) features to extract long-term spatial structural information from historical edge images, enhancing the model’s interpretability. The proposed method achieves an accuracy of 99.187% on a self-constructed dataset. Additionally, it attains a real-time inference speed of 20.983 ms per sample on a hardware platform equipped with an Intel i9-12900H CPU and an RTX 3060 GPU, thus effectively balancing accuracy, speed, and interpretability. Full article

This study examined the effects of rearing, physiological, and processing conditions on the volatile profile of Atlantic salmon. Fish were reared under two different temperature and light conditions, and three harvests were conducted at different time points for male and female fish. Fish were processed to yield fillets with or without skin. Volatiles were analyzed using SIFT-MS headspace analysis. Atlantic salmon reared in cooler temperatures under a 12 h light/dark cycle exhibited significantly lower concentrations of off-odor volatiles compared to those reared in warm conditions under continuous light, suggesting that cooler temperatures with a dark cycle help maintain freshness. A temperature shift from cool to warm further increased volatile accumulation. Longer rearing time resulted in higher volatile concentrations, attributed to greater biochemical products, increased susceptibility to lipid oxidation, protein degradation, and contaminant accumulation from the rearing environment. Males had higher volatile levels at 202 days, while females surpassed males by 242 days, likely due to increased biochemical accumulation associated with reproductive development. Fillets with skin exhibited significantly higher concentration of off-odor volatiles. These findings highlight the role of all studied factors in establishing optimum conditions to minimize spoilage-related volatiles and preserve the freshness of Atlantic salmon, with rearing temperature being the most critical factor. Full article

Despite significant advances in deep learning-based image captioning, many state-of-the-art models still struggle to balance visual grounding (i.e., accurate object and scene descriptions) with linguistic coherence (i.e., grammatical fluency and appropriate use of non-visual tokens such as articles and prepositions). To address these limitations, we propose a hybrid image captioning framework that integrates handcrafted and deep visual features. Specifically, we combine local descriptors—Scale-Invariant Feature Transform (SIFT) and Bag of Features (BoF)—with high-level semantic features extracted using ResNet50. This dual representation captures both fine-grained spatial details and contextual semantics. The decoder employs Bahdanau attention refined with an Attention-on-Attention (AoA) mechanism to optimize visual-textual alignment, while GloVe embeddings and a GRU-based sequence model ensure fluent language generation. The proposed system is trained on 200,000 image-caption pairs from the MS COCO train2014 dataset and evaluated on 50,000 held-out MS COCO pairs plus the Flickr8K benchmark. Our model achieves a CIDEr score of 128.3 and a SPICE score of 29.24, reflecting clear improvements over baselines in both semantic precision—particularly for spatial relationships—and grammatical fluency. These results validate that combining classical computer vision techniques with modern attention mechanisms yields more interpretable and linguistically precise captions, addressing key limitations in neural caption generation. Full article

Tea is one of the most consumed beverages in the world and presents a great aromatic diversity depending on the origin of the production and the transformation process. Volatile organic compounds (VOCs) greatly contribute to the sensory perception of tea and are excellent markers for traceability and quality. In this work, we analyzed the volatile organic compounds (VOCs) emitted by twenty-six perfectly traced samples of tea with two analytical techniques and two data treatment strategies. First, we performed headspace solid-phase microextraction gas chromatography–mass spectrometry (HS-SPME-GC-MS) as the most widely used reference method for sanitary and quality controls of food. Next, we analyzed the samples with selected-ion flow-tube mass spectrometry (SIFT-MS), an emerging method for direct analysis of food products and aroma. We compared the performances of both techniques to trace the origin and the transformation processes. We selected the forty-eight most relevant markers with HS-SPME-GC-MS and evaluated their concentrations with a flame ionization detector (FID) on the same instrument. This set of markers permitted separation of the origins of samples but did not allow the samples to be differentiated based on the color. The same set of markers was measured with SIFT-MS instrument without success for either origin separation or color differentiation. Finally, a post-processing treatment of raw data signals with an untargeted approach was applied to the GC-MS and SIFT-MS dataset. This strategy allowed a good discrimination of origin and color with both instruments. Advantages and drawbacks of volatile profiles with both instruments were discussed for the traceability and quality assessment of food. Full article

Processing scenery and finding points of interest is crucial for applications in robotics and aerospace missions. Those areas require efficient and reliable visual input processing. Here, field programmable gate arrays (FPGAs) offer essential advantages, like low power consumption compared to CPUs, performing a large number of calculations simultaneously, and having compact hardware. This paper presents an FPGA system that processes incoming camera data, finds points of interest, and matches them across different images on high-resolution images (2048 × 1088). It is a novel approach to implement the complete image processing pipeline on high-resolution images within the FPGA fabric without additional hardware. For keypoint detection and matching, our work uses a modified SIFT algorithm optimized for FPGA implementation processing and a nearest neighbor-based matching method. It was implemented on a Xilinx Kintex-7 FPGA and partially on a NanoXplore NG-Ultra to evaluate a radiation-hardened FPGA for space applications. On the Kintex-7, the keypoint detection achieves a speed of 33 ms per image, and its features are matched on up to 5 images per second. Judging by the resource utilization of one image processing module on the NG-Ultra, porting the entire system on a radiation-hardened FPGA appears feasible. Full article

Magnetic resonance sounding is a geophysical method for quantitatively determining the state for groundwater storage that has gained international attention in recent years. However, the practical acquisition of magnetic resonance sounding signals, which are on the nanovolt scale, is susceptible to various types of interference, such as power-line harmonics, random noise, and spike noise. Such interference can degrade the quality of magnetic resonance sounding signals and, in severe cases, be completely drowned out by noise. This paper introduces an adaptive Gaussian filtering algorithm that is well-suited for handling intricate noise signals due to its adaptive solving characteristics and iterative sifting approach. Notably, the algorithm can process signals without relying on prior knowledge. The adaptive Gaussian filtering algorithm is applied for the envelope extraction of noisy magnetic resonance sounding signals, and the reliability and effectiveness of the method are rigorously validated. The simulation results reveal that, even under strong noise interference (with original signal-to-noise ratios ranging from −7 dB to −25 dB), the magnetic resonance sounding signal obtained after algorithmic processing is compared to the ideal signal, with 16 sets of data statistics, and the algorithm ensures an initial amplitude uncertainty within 4nV and restricts the uncertainty of the relaxation time within a 6 ms range. The signal-to-noise ratio can be boosted by up to 53 dB. The comparative assessments with classical algorithms such as empirical mode decomposition and the harmonic modeling method confirm the superior performance of the adaptive Gaussian filtering algorithm. The processing of the field data also fully proved the practical application effects of the algorithm. Full article

Personal care products (PCPs) are intended for regular application by consumers and therefore assuring the safety of these products is very important. Recently, benzene contamination has been highlighted in certain PCPs. The present study applies selected ion flow tube mass spectrometry (SIFT-MS) to a simultaneous headspace analysis of benzene, 1,4-dioxane, and formaldehyde—all known or suspected carcinogens—in nine haircare products with supporting qualitative analysis by gas chromatography–mass spectrometry (GC-MS). Headspace-SIFT-MS method development is compatible with the method of standard additions, which is necessary for the quantitation of volatile impurities in these complex emulsions. Benzene was quantified above the low-ng g⁻¹ limit of quantitation (LOQ) in three products, dioxane above the sub-μg g⁻¹ LOQ in all products, and formaldehyde above the low-μg g⁻¹ LOQ in two products, providing a quantitative analysis at concentrations relevant to consumer safety. This study facilitated the development of generic workflows for SIFT-MS method development and application in routine analysis of PCPs. The assessment of workflows for SIFT-MS compared to a conventional GC-MS analysis suggests that 8- to 30-fold throughput enhancements may be possible for quantitative and screening analysis using SIFT-MS. Full article

Since the initial 2018 recall of angiotensin receptor blockers due to unacceptable levels of mutagenic N-nitrosodimethylamine (NDMA) impurity, numerous drug products delivering diverse active pharmaceutical ingredients (APIs) have been recalled. Regulators and the industry are working together to understand and address this widescale problem. Conventional analysis of NDMA utilizes liquid or gas chromatography-based procedures that can involve complicated sample preparation and slow sample analysis. Selected ion flow tube mass spectrometry (SIFT-MS) analyses NDMA directly in the gas phase using soft chemical ionization, with an LOQ of 2 ng g⁻¹. Through the novel application of the multiple headspace extraction (MHE) technique, NDMA was quantified directly and rapidly from the drug product without dissolution, at levels well below the regulatory acceptable intake of 96 ng day⁻¹. A comparative analysis of recalled metformin using MHE-SIFT-MS and a conventional liquid chromatography–mass spectrometry/mass spectrometry (LC-MS/MS) method showed good agreement. Use of the novel MHE-SIFT-MS approach may enable a wider screening of drug products to be conducted, since it provides around a three-fold increase in daily sample throughput. Full article

This article is focused on the comprehensive evaluation of alleyways to scale-invariant feature transform (SIFT) and random sample consensus (RANSAC) based multispectral (MS) image registration. In this paper, the idea is to extensively evaluate three such SIFT- and RANSAC-based registration approaches over a heterogenous mix containing Triticum aestivum crop and Raphanus raphanistrum weed. The first method is based on the application of a homography matrix, derived during the registration of MS images on spatial coordinates of individual annotations to achieve spatial realignment. The second method is based on the registration of binary masks derived from the ground truth of individual spectral channels. The third method is based on the registration of only the masked pixels of interest across the respective spectral channels. It was found that the MS image registration technique based on the registration of binary masks derived from the manually segmented images exhibited the highest accuracy, followed by the technique involving registration of masked pixels, and lastly, registration based on the spatial realignment of annotations. Among automatically segmented images, the technique based on the registration of automatically predicted mask instances exhibited higher accuracy than the technique based on the registration of masked pixels. In the ground truth images, the annotations performed through the near-infrared channel were found to have a higher accuracy, followed by green, blue, and red spectral channels. Among the automatically segmented images, the accuracy of the blue channel was observed to exhibit a higher accuracy, followed by the green, near-infrared, and red channels. At the individual instance level, the registration based on binary masks depicted the highest accuracy in the green channel, followed by the method based on the registration of masked pixels in the red channel, and lastly, the method based on the spatial realignment of annotations in the green channel. The instance detection of wild radish with YOLOv8l-seg was observed at a mAP@0.5 of 92.11% and a segmentation accuracy of 98% towards segmenting its binary mask instances. Full article

The existing image matching methods for remote sensing scenes are usually based on local features. The most common local features like SIFT can be used to extract point features. However, this kind of methods may extract too many keypoints on the background, resulting in low attention to the main object in a single image, increasing resource consumption and limiting their performance. To address this issue, we propose a method that could be implemented well on resource-limited satellites for remote sensing images ship matching by leveraging line features. A keypoint extraction strategy called line feature based keypoint detection (LFKD) is designed using line features to choose and filter keypoints. It can strengthen the features at corners and edges of objects and also can significantly reduce the number of keypoints that cause false matches. We also present an end-to-end matching process dependent on a new crop patching function, which helps to reduce complexity. The matching accuracy achieved by the proposed method reaches 0.972 with only 313 M memory and 138 ms testing time. Compared to the state-of-the-art methods in remote sensing scenes in extensive experiments, our keypoint extraction method can be combined with all existing CNN models that can obtain descriptors, and also improve the matching accuracy. The results show that our method can achieve ∼50% test speed boost and ∼30% memory saving in our created dataset and public datasets. Full article

The pollution of air and water with volatile organic compounds (VOCs), both hazardous and odorous, is of significant concern due to impacts on human health and quality of life, as well as the environment. South Korea is a highly industrialized and densely populated nation and suffers from significant VOC and ozone pollution. In recent years, South Korea has implemented more stringent controls on industry to address air and water pollution, requiring more real-time and on-site analysis. The selected ion flow tube mass spectrometry (SIFT-MS) technique has been increasingly adopted to monitor source emissions and their dispersion, enabling a more rapid response to pollution incidents. To this end, the flexibility of SIFT-MS instrumentation for both laboratory- and field-based analysis, including in mobile laboratories, has been valuable. SIFT-MS has been applied to emission source characterization, fenceline monitoring, ambient monitoring, pollution mapping, and incident response (including the use of drone-based sampling) for hazardous air pollutants (HAPs), odor nuisance species, and compounds that have high ozone formation potential (OFP) and/or contribute to secondary aerosol (SOA) formation. This review summarizes the South Korean application of SIFT-MS to the monitoring of VOC pollutants. Full article

Sunflower seeds are a popular snack in many countries, such as the United States, China, and Spain. Sunflower seeds are typically roasted to create desirable aromas before being eaten. The desirable aromas are created by the Maillard and lipid oxidation reactions. Increasing the volatiles created by these reactions can create a more desirable product, increasing consumer acceptance of sunflower seeds. Seeds were soaked in solutions at pH 4, 7, and 9 and with added glucose, fructose, whey protein isolate, or whey protein concentrate before roasting. The resulting seeds were evaluated by selected-ion flow tube mass spectrometry to determine the volatile concentrations and by an untrained panel of consumers to determine acceptability. Increasing the pH increased the pyrazines but did not affect other volatiles. Adding reducing sugars or whey protein increased most volatiles. The fructose increased dimethylpyrazines, 2-methylpyrazine, and trimethylpyrazine concentrations more than glucose. However, the glucose increased furfural concentration more than fructose. The whey protein concentrate increased volatile levels more than any other treatment. The total Maillard volatiles and Browning index were increased by the same treatments. Sensory indicated that fructose increased desirable aroma the most, followed by whey protein treatments, and both were liked more than the pH 7 control. Optimizing roasting conditions by increasing the pH and reducing sugar and protein content can favor the Maillard reaction conditions, increasing the positive aromas associated with roasted sunflower seeds. Full article

This study addresses the growing need for phytoncide studies, driven by the demand to design indoor forest healing programs, including virtual reality experiences, for patients unable to visit actual forests. Previous studies have struggled to establish consistent phytoncide emission patterns in outdoor forest environments owing to varying microclimates and abiotic factors. In addition, the traditional gas chromatography–mass spectroscopy (GC-MS) method presents field measurement challenges, whereas the selected ion flow tube (SIFT)-MS method offers improved efficiency. This study concentrated on a controlled phytotron environment and compared the GC-MS and SIFT-MS findings, revealing similar emission trends with slightly higher SIFT-MS concentrations. Daily phytoncide emissions fluctuated with light intensity and abiotic stressors. Both methods consistently detected pinenes, primarily emitted by Pinus strobus L. seedlings, in the phytotron. Statistical analysis confirmed the compatibility between GC-MS and SIFT-MS results, supporting the use of SIFT-MS for forest phytoncide assessment. In the second phase, the phytoncide emissions were assessed indoors, outdoors, and in the phytotron, highlighting the superiority of the phytotron under controlled conditions. Despite certain limitations, this study underscores the value of phytotron-based measurements for indoor forest healing programs and the potential adoption of SIFT-MS in future field-based phytoncide research. Full article

Conventional gas chromatography-mass spectrometry (GC-MS) analysis methods for measuring the concentration of phytoncides in forests are limited because of the need for an extended human presence in forests, the risk of errors, and contamination. To overcome these issues, this study introduces real-time measurement devices and selected ion flow tube mass spectrometry (SIFT-MS) as potential replacements. This study was conducted in the Hongneung Experimental Forest between 19 and 21 November 2019. A correlation analysis and independent samples t-test were performed to compare the GC-MS and SIFT-MS techniques. The diurnal patterns and trends in the phytoncide concentrations analyzed using the GC-MS and SIFT-MS methods were similar, suggesting the potential replacement of GC-MS with SIFT-MS. While both methods revealed similar major components in the daytime nonvolatile organic compounds (NVOCs), with pinenes comprising approximately half of the total percentage, the 24 h SIFT-MS analysis indicated reduced proportions of pinenes and benzaldehyde, along with the detection of more diverse NVOC compounds at night. Additionally, the studies indicated that GC-MS exhibited slightly higher selectivity, resulting in the detection of fewer NVOC compounds with SIFT-MS. The correlation analysis between the microclimate indicators and phytoncide measurement methods revealed differences: GC-MS with a mini pump showed positive correlations with fine dust and industrially derived VOCs, while the 24 h real-time measurements exhibited strong negative correlations. Consequently, while the GC-MS and SIFT-MS methods exhibited both similarities and differences in phytoncide concentrations, an independent samples t-test, confirming no statistically significant differences between the two methods, suggests the suitability of adopting SIFT-MS over GC-MS for phytoncide collection and analysis in forest environments. Nevertheless, this study contributes to the literature by comparing outdoor phytoncide levels using the GC-MS and SIFT-MS methodologies. These findings, which show that the methods are closely aligned, can guide future researchers in considering SIFT-MS equipment for phytoncide studies, offering a more accessible and efficient option with real-time capabilities. Full article

Many rivers are polluted by trash and garbage that can affect the environment. Riverbank inspection usually relies on workers of the environmental protection office, but sometimes the places are unreachable. This study applies unmanned aerial vehicles (UAVs) to perform the inspection task, which can significantly relieve labor work. Two UAVs are used to cover a wide area of riverside and capture riverbank images. The images from different UAVs are stitched using the scale-invariant feature transform (SIFT) algorithm. Static and dynamic image stitching are tested. Different you only look once (YOLO) algorithms are applied to identify riverbank garbage. Modified YOLO algorithms improve the accuracy of riverine waste identification, while the SIFT algorithm stitches the images obtained from the UAV cameras. Then, the stitching results and garbage data are sent to a video streaming server, allowing government officials to check waste information from the real-time multi-camera stitching images. The UAVs utilize 4G communication to transmit the video stream to the server. The transmission distance is long enough for this study, and the reliability is excellent in the test fields that are covered by the 4G communication network. In the automatic reconnection mechanism, we set the timeout to 1.8 s. The UAVs will automatically reconnect to the video streaming server if the disconnection time exceeds the timeout. Based on the energy provided by the onboard battery, the UAV can be operated for 20 min in a mission. The UAV inspection distance along a preplanned path is about 1 km at a speed of 1 m/s. The proposed UAV system can replace inspection labor, successfully identify riverside garbage, and transmit the related information and location on the map to the ground control center in real time. Full article