Search

Weeds are mainly spread by weed seeds being mixed with agricultural and forestry crop seeds, grain, animal hair, and other plant products, and disturb the growing environment of target plants such as crops and wild native plants. The accurate and efficient classification of weed seeds is important for the effective management and control of weeds. However, classification remains mainly dependent on destructive sampling-based manual inspection, which has a high cost and rather low flux. We considered that this problem could be solved using a nondestructive intelligent image recognition method. First, on the basis of the establishment of the image acquisition system for weed seeds, images of single weed seeds were rapidly and completely segmented, and a total of 47 696 samples of 140 species of weed seeds and foreign materials remained. Then, six popular and novel deep Convolutional Neural Network (CNN) models are compared to identify the best method for intelligently identifying 140 species of weed seeds. Of these samples, 33 600 samples are randomly selected as the training dataset for model training, and the remaining 14 096 samples are used as the testing dataset for model testing. AlexNet and GoogLeNet emerged from the quantitative evaluation as the best methods. AlexNet has strong classification accuracy and efficiency (low time consumption), and GoogLeNet has the best classification accuracy. A suitable CNN model for weed seed classification could be selected according to specific identification accuracy requirements and time costs of applications. This research is beneficial for developing a detection system for weed seeds in various applications. The resolution of taxonomic issues and problems associated with the identification of these weed seeds may allow for more effective management and control.

We propose a locality preserving partial least squares discriminant analysis (LPPLSDA) which adds a locality preserving feature to the conventional partial least squares discriminant analysis(PLS-DA). The locality preserving feature captures the within group structural information via a similarity graph. The ability of LPPLS-DA to capture local structures allows it to be better suited for face recognition. We evaluate the performance of our proposed method on several benchmarked face databases which offer different levels of complexity in terms of sample size as well as image acquisition conditions. The experimental results indicate that, for each database used, the proposed method consistently outperformed the conventional PLS-DA method.

The widespread Pomacea canaliculata and Pomacea maculata in North America and Asia have caused significant adverse effects on the local ecological environment and residents' health. Timely knowledge of the distribution of eggs from the two Pomacea spp. in a certain region can effectively reduce the cost of treatment and improve prevention effectiveness. Most of the existing methods are only able to identify eggs from the two Pomacea spp. or detect them in specific but not natural environments while they cannot achieve good results in the face of a complex real-world scene. This letter proposes a model for detecting eggs from the two Pomacea spp. based on dynamic convolution and multiscale feature fusion. The model can identify and locate the eggs of the two Pomacea spp. effectively. At the same time, we combined the proposed model with scale invariant feature transform (SIFT) algorithm to design a system for counting eggs of the two Pomacea spp., which can automatically identify the eggs in the actual natural environment and alleviate duplicate counting caused by image acquisition. Besides, we also built a dataset of 20,000 images of Pomacea canaliculata eggs and Pomacea maculata eggs from unmanned aerial vehicle (UAV) aerial photography. Experimental results showed that the proposed deep learning model has a better performance than others, and the proposed computer vision system can be successfully applied to support Pomacea spp. disease management.

The symptoms associated with deficiencies in plants tends to appear often on the leaves. The color and shape of a leaf often used for diagnosing the nutritional deficiencies in plants and classification of these properties often pose serious problem. Since same color and shape may have many root cause problem. It is hence necessary to carefully analyse the texture of leaf with proper training of a classifier. In this paper, we design an image acquisition based classification model that utilizes Internet of Things (IoTs) for data acquisition and a variant of convolutional neural networks (CNN) namely DenseNet-BC for the task of classification. Prior classification, the model is trained over several iteration based on careful observation of features and its related symptoms. The simulation is conducted with fine-tuning of classification after several iterations. The results of simulation shows that the proposed method obtains improved classification accuracy in terms of accuracy and F-measure than other deep learning models.

Whole slide imaging (WSI) has been increasingly adopted for digital evaluation of surgical pathology specimens. Unlike histological slides, cytological preparations frequently display a heterogeneous distribution of cells throughout slides in different focal planes sometimes admixed with obscuring material, therefore requiring multiple scanning planes which significantly lengthens image acquisition and evaluation times. Although examination of digital images can be more advantageous than conventional glass slides, the challenges of focusing, scanning and screening cytological specimens and the associated increase in scan times and data storage needs have limited the routine application of WSI in cytopathology practice. Emerging digital systems designed to overcome image acquisition obstacles coupled with artificial intelligence algorithms augmenting screening of digital cytology slides offer innovative solutions to address these limitations. The aim of this review is to critically address the potential benefits and pitfalls of employing WSI in cytopathology practice and to introduce promising state-of-the-art solutions on the horizon.

In order to pay more attention to the quality of construction concrete and accurately judge whether concrete material meets the standard, a nondestructive testing algorithm of building concrete mat...

Advances in medical imaging technologies now allow noninvasive image acquisition from individual patients at high spatiotemporal resolutions. A relatively new effort of predictive oncology is to develop a paradigm for forecasting the future status of an individual tumor given initial conditions and an appropriate mathematical model. The objective of this study was to introduce a comprehensive multiscale computational method to predict cancer and microvascular network growth patterns. A rectangular lattice-based model was designed so different evolutionary scenarios could be simulated and for predicting the impact of diffusible factors on tumor morphology and size. Further, the model allows prediction-based simulation of cell and microvascular behavior. Like a single cell, each agent is fully realized within the model and interactions are governed in part by machine learning methods. This multiscale computational model was developed and incorporated input information from in vivo microscale computed tomography (microCT) images acquired from breast cancer-bearing mice. It was found that as the difference between expansion of the cancer cell population and microvascular network increases, cells undergo proliferation and migration with a greater probability compared to other phenotypes. Overall, multiscale computational model agreed with both theoretical expectations and experimental findings (microCT images) not used during model training.

In order to solve the problem of crop disease detection in large-scale planting, a new crop disease detection algorithm based on multi-feature decision fusion is proposed. This paper proposes a multi-feature decision fusion disease discrimination algorithm (PD R-CNN) based on machine vision on crop surfaces. The algorithm is based on the machine vision processing model of R-CNN and integrates a disease discrimination algorithm on the basis of R-CNN. After training on crop image data sets, PD R-CNN can reach the goal of identifying crop surface lesions. This paper uses machine vision image acquisition, image processing and analysis technology to collect and analyze the growth of cucumber seedlings. The research results show that compared with manual judgment, PD R-CNN reduces the workload and can effectively distinguish crop diseases. Through experiments, during the occurrence of pests and diseases, PD R-CNN has a monitoring accuracy of 88.0% for mosaic disease, 92.0% for root rot, 88.0% for powdery mildew, and 86.0% for aphids, indicating that there are errors in actual monitoring, but the accuracy exceeds 85.0% can be put into use.

Computed tomography is the most commonly used imaging modality to detect and stage pancreatic cancer. Previous advances in pancreatic cancer imaging have focused on optimizing image acquisition parameters and reporting standards. However, current state-of-the-art imaging approaches still misdiagnose some potentially curable pancreatic cancers and do not provide prognostic information or inform optimal management strategies beyond stage. Several recent developments in pancreatic cancer imaging, including artificial intelligence and advanced visualization techniques, are rapidly changing the field. The purpose of this article is to review how these recent advances have the potential to revolutionize pancreatic cancer imaging.

Automating the extraction of meaningful temporal information from sequences of microscopy images represents a major challenge to characterize dynamical biological processes. So far, strong limitations in the ability to quantitatively analyze single-cell trajectories have prevented large-scale investigations to assess the dynamics of entry into replicative senescence in yeast. Here, we have developed DetecDiv, a microfluidic-based image acquisition platform combined with deep learning-based software for high-throughput single-cell division tracking. We show that DetecDiv can automatically reconstruct cellular replicative lifespans with high accuracy and performs similarly with various imaging platforms and geometries of microfluidic traps. In addition, this methodology provides comprehensive temporal cellular metrics using time-series classification and image semantic segmentation. Last, we show that this method can be further applied to automatically quantify the dynamics of cellular adaptation and the real-time cell survival upon exposure to environmental stress. Hence, this methodology provides an all-in-one toolbox for high-throughput phenotyping for cell cycle, stress response, and replicative lifespan assays.

Deep learning is an idea technique for image classification. Imaging flow cytometer enables high throughput cell image acquisition and some have integrated with real-time cell sorting. The combination of deep learning and imaging flow cytometer has changed the landscape of high throughput cell analysis research. In this review, we focus on deep learning technologies applied in imaging flow cytometer for cell classification and real-time cell sorting. This article describes some recent research, challenges and future trend in this area.

Synopsis Digital photography and videography provide rich data for the study of animal behavior and are consequently widely used techniques. For fixed, unmoving cameras there is a resolution versus field-of-view tradeoff and motion blur smears the subject on the sensor during exposure. While these fundamental tradeoffs with stationary cameras can be sidestepped by employing multiple cameras and providing additional illumination, this may not always be desirable. An alternative that overcomes these issues of stationary cameras is to direct a high-magnification camera at an animal continually as it moves. Here, we review systems in which automatic tracking is used to maintain an animal in the working volume of a moving optical path. Such methods provide an opportunity to escape the tradeoff between resolution and field of view and also to reduce motion blur while still enabling automated image acquisition. We argue that further development will be useful and outline potential innovations that may improve the technology and lead to more widespread use.

3D digital reconstruction techniques are extensively used for quality control purposes. Among them, photogrammetry and photometric stereo methods have been for a long time used with success in several application fields. However, generating highly-detailed and reliable micro-measurements of non-collaborative surfaces is still an open issue. In these cases, photogrammetry can provide accurate low-frequency 3D information, whereas it struggles to extract reliable high-frequency details. Conversely, photometric stereo can recover a very detailed surface topography, although global surface deformation is often present. In this paper, we present the preliminary results of an ongoing project aiming to combine photogrammetry and photometric stereo in a synergetic fusion of the two techniques. Particularly, hereafter, we introduce the main concept design behind an image acquisition system we developed to capture images from different positions and under different lighting conditions as required by photogrammetry and photometric stereo techniques. We show the benefit of such a combination through some experimental tests. The experiments showed that the proposed method recovers the surface topography at the same high-resolution achievable with photometric stereo while preserving the photogrammetric accuracy. Furthermore, we exploit light directionality and multiple light sources to improve the quality of dense image matching in poorly textured surfaces.

In recent years, with the development of communication technology, embedded computing technology and sensor technology, it has become increasingly mature. Micro sensors with sensing, computing and communication capabilities have appeared in large numbers and developed rapidly, making wireless sensor networks widely used. People put forward higher requirements for the accuracy, reliability and flexibility of the image acquisition system. The image transmission system using analog technology not only has low image quality, but also has a serious waste of system resources, is not easy to form a complex network structure, and has poor functional scalability. In view of the actual needs of the current image acquisition and wireless transmission system, based on embedded technology, image acquisition, processing technology and network transmission technology, this paper proposes and designs a low-cost, high-reliability embedded image acquisition and wireless transmission system. Experimental tests show that this system has reasonable design, high video coding efficiency, good image continuity, stable operation, and basically realizes the display, storage and playback functions of the collected video data. Improve the transmission rate of the system and reduce the distortion caused by compression in terms of image compression. At the same time, it supports multiple image resolutions, frame rate options and multiple video formats, and the system’s transmission rate can adapt to the state of the network. This design fulfills the basic requirements of an embedded image acquisition system based on network technology, and provides a good foundation for the next development of a gigabit network-based image acquisition system.

In this paper, a robust visual servoing control approach is proposed to address the landing problem for quadrotors on a moving platform. A vision system is implemented to estimate the position and velocity of the quadrotor. A robust cascade controller is proposed by following backstepping-like fundamentals and robust compensating theory. The effects of time-varying uncertainties, including parameter uncertainties and external disturbances, and time-varying delays resulted from image acquisition, image processing, and sensor measurement delays can be restrained. Experimental results using a quadrotor to land on a ground moving target illustrate the effectiveness of the proposed approach.

Indonesia is a country with high rice needs because it is a staple food for more than 90% of populations. High demand requires high stock so imports are carried out in accordance with Permendagri Number 19/M-DAG/PER/3/2014 which explains rice import standards. There are many types of rice imported into Indonesia with various quality, color and import requirements such as for health or price stabilization. In terms of colors, imported white rice is the most consumed rice by Indonesians. One example is jasmine rice from Thailand. Meanwhile, in terms of imports, both for health and stabilizing the price of japonica rice (Japan) and Basmati (Pakistan) are the most imported to Indonesia. But there are still many who are not familiar with those three rices. In this research, the three types of rice were identified by comparing the Multi-SVM algorithm and Neural Network algorithm. Image acquisition is done using a flatbed scanner which produces 90 images divided into 63 training images and 27 testing images. K-Means becomes an image segmentation method and image binary converts. Feature extraction using morphological features with the regionprop method combined with the Gray Level Co-Occence Matrix (GLCM) produces 9 features that can produce 96.296% accuracy for Multi-SVM and 88.89% Neural Network, Indonesia merupakan negara dengan kebutuhan beras yang tinggi karena menjadi makanan pokok lebih dari 90% masyarakatnya. Kebutuhan tinggi memerlukan stok yang tinggi sehingga impor pun dilakukan dengan aturan-aturan sesuai Permendagri Nomor 19/M-DAG/PER/3/2014 yang menerangkan standar impor beras. Terdapat banyak jenis beras yang di impor ke Indonesia dengan berbagai kualitas, warna dan keperluan impor seperti untuk kesehatan atau stabilisasi harga. Dari sisi wana, beras putih impor merupakan beras paling banyak dikonsumsi oleh warga Indonesia. Salahsatu contohnya adalah beras melati (jasmine) dari Thailand. Sedangkan dari sisi keperluan impor baik untuk kesehatan dan stabilisasi harga beras japonica (Jepang) dan Basmati (Pakistan) merupakan jenis paling banyak di impor ke Indonesia. Namun masih banyak yang belum mengenal ketiga beras tersebut. Pada penelitian ini dilakukan identifikasi ketiga jenis beras tersebut dengan perbandingan algoritma Multi-SVM dan Neural Network. Akuisisi citra dilakukan menggunakan flatbed scanner yang menghasilkan 90 citra yang dibagi menjadi 63 citra training dan 27 citra testing. K-Means menjadi metode segmentasi citra dan mengubah citra menjadi citra biner. Ekstraksi fitur menggunakan morphological feature dengan metode regionprop dikombinasikan dengan Gray Level Co-Occence Matrix (GLCM) menghasilkan 9 fitur yang dapat menghasilkan akurasi 96,296% untuk Multi-SVM dan Neural Network 88,89%.

Iris recognition used the iris features to verify and identify the identity of human. The iris has many advantages such as stability over time, easy to use and high recognition accuracy. However, the poor quality of iris images can degrade the recognition accuracy of iris recognition system. The recognition accuracy of this system is depended on the iris pattern quality captured during the iris acquisition. The iris pattern quality can degrade due to the blurry image. Blurry image happened due to the movement during image acquisition and poor camera resolution. Due to that, a deblurring method based on the Wiener filter was proposed to improve the quality of iris pattern. This work is significant since the proposed method can enhance the quality of iris pattern in the blurry image. Based to the results, the proposed method improved the quality of iris pattern in the blurry image. Moreover, it recorded the fastest execution time to improve the quality of iris pattern compared to the other methods.

As the COVID-19 pandemic impacts global populations, computed tomography (CT) lung imaging is being used in many countries to help manage patient care as well as to rapidly identify potentially useful quantitative COVID-19 CT imaging biomarkers. Quantitative COVID-19 CT imaging applications, typically based on computer vision modeling and artificial intelligence algorithms, include the potential for better methods to assess COVID-19 extent and severity, assist with differential diagnosis of COVID-19 versus other respiratory conditions, and predict disease trajectory. To help accelerate the development of robust quantitative imaging algorithms and tools, it is critical that CT imaging is obtained following best practices of the quantitative lung CT imaging community. Toward this end, the Radiological Society of North America's (RSNA) Quantitative Imaging Biomarkers Alliance (QIBA) CT Lung Density Profile Committee and CT Small Lung Nodule Profile Committee developed a set of best practices to guide clinical sites using quantitative imaging solutions and to accelerate the international development of quantitative CT algorithms for COVID-19. This guidance document provides quantitative CT lung imaging recommendations for COVID-19 CT imaging, including recommended CT image acquisition settings for contemporary CT scanners. Additional best practice guidance is provided on scientific publication reporting of quantitative CT imaging methods and the importance of contributing COVID-19 CT imaging datasets to open science research databases.

Biometric applications are very sensitive to the process because of its complexity in presenting unstructured input to the processing. The existing applications of image processing are based on the implementation of different programing segments such as image acquisition, segmentation, extraction, and final output. The proposed model is designed with 2 convolution layers and 3 dense layers. We examined the module with 5 datasets including 3 benchmark datasets, namely CASIA, UBIRIS, MMU, random dataset, and the live video. We calculated the FPR, FNR, Precision, Recall, and accuracy of each dataset. The calculated accuracy of CASIA using the proposed system is 82.8%, for UBIRIS is 86%, MMU is 84%, and the random dataset is 84%. On live video with low resolution, calculated accuracy is 72.4%. The proposed system achieved better accuracy compared to existing state-of-the-art systems.

Microscopy inspection of the morphology of activated sludge (AS) flocs can provide important information regarding the AS properties, which strongly affect the performance of AS systems. However, the acquisition of such information from microscopy inspection results requires skilled and specialized expertise. In this study, we aimed to develop two deep learning-based two-label classifiers for recognizing aggregated or dispersed flocs (classifier-1) and the presence or absence of filamentous bacteria (classifier-2). To achieve this, we used a convolutional neural network (CNN)-based method and selected the pre-trained Inception v3 as the CNN architecture. We developed an automatic microscopy image acquisition system, enabling us to obtain 154 images for 7 min. Over 12 000 images of aggregated and dispersed flocs were obtained from wastewater treatment plant (WWTP)-S and -E over 15 weeks. Classifier-1 was retrained using these images. Classifier-1 distinguished the aggregated and dispersed flocs with a training accuracy of approximately 95% and recognized a 20% morphological change in the aggregated flocs. Classifier-1 also recognized the morphology of AS flocs obtained from other WWTPs, the AS from which was used for retraining. Classifier-2 quantitatively recognized an abundance of filamentous bacteria in the AS flocs. These results clearly indicated that the developed image classification model could serve as a useful warning system for the settleability deterioration and abundance of filamentous bacteria in the aeration tank of a full-scale AS system.

Purpose In MRgRT, accuracy of treatment depends on the gating latency, when real-time targeting and gating is enabled. Gating latency is dependent on image acquisition, processing time, accuracy, efficacy of target tracking algorithms, and radiation beam delivery latency. In this report, clinical experience of the MRI4D QUASAR motion phantom for latency measurements on a 0.35-T magnetic resonance-linear accelerator (MR-LINAC) with two imaging speeds and four tracking algorithms was studied. Materials/methods Beam-control latency was measured on a 0.35-T MR-LINAC system with four target tracking algorithms and two real-time cine imaging sequences [four and eight frames per second (FPS)]. Using an MR-compatible motion phantom, the delays between phantom beam triggering signal and linac radiation beam control signal were evaluated for three motion periods with a rigid target. The gating point was set to be 8 mm above the full exhalation position. The beam-off latency was measured for a total of 24 combinations of tracking algorithm, imaging FPS, and motion periods. The corresponding gating target margins were determined using the target motion speed multiplied by the beam-off latency. Results The largest measured beam-off latency was 302 ± 20 ms with the Large Deforming Targets (LDT) algorithm and 4 s motion period imaged with 8-FPS cine MRI. The corresponding gating uncertainty based on target motion speed was 3.0 mm. The range of the average beam-off latency was 128-243 ms in 4-FPS imaging and 47-302 ms in 8-FPS imaging. Conclusions The gating latency was measured using an MRI4D QUASAR motion phantom in a 0.35-T MR-LINAC. The latency measurements include time delay related to MR imaging method, target tracking algorithm and system delay. The gating uncertainty was estimated based on the beam-off latency measurements and the target motion.

The morphological features of fish, such as the body length, the body width, the caudal peduncle length, the caudal peduncle width, the pupil diameter, and the eye diameter are very important indicators in smart mariculture. Therefore, the accurate measurement of the morphological features is of great significance. However, the existing measurement methods mainly rely on manual measurement, which is operationally complex, low efficiency, and high subjectivity. To address these issues, this paper proposes a scheme for segmenting fish image and measuring fish morphological features indicators based on Mask R-CNN. Firstly, the fish body images are acquired by a home-made image acquisition device. Then, the fish images are preprocessed and labeled, and fed into the Mask R-CNN for training. Finally, the trained model is used to segment fish image, thus the morphological features indicators of the fish can be obtained. The experimental results demonstrate that the proposed scheme can segment the fish body in pure and complex backgrounds with remarkable performance. In pure background, the average relative errors (AREs) of all indicators measured all are less than 2.8%, and the AREs of body length and body width are less than 0.8%. In complex background, the AREs of all indicators are less than 3%, and the AREs of body length and body width is less than 1.8%.

Appearance preservation aims to estimate reflectance functions to model the way real materials interact with light. These functions are especially useful in digital preservation of heritage and realistic rendering, as they reproduce the appearance of real materials in virtual scenes. This work proposes an image-based process that aims to preserve the appearance of surfaces whose reflectance properties are spatially variant. During image acquisition, this process considers the whole environment as a source of light over the area to be preserved and, assuming the environment is static, it does not require controlled environments. To achieve this goal, the scene geometry and relative camera positions are approximated from a set of HDR images taken inside the real scene, using a combination of structure from motion and multi-view stereo methods. Based on this data, a set of unstructured lumigraphs is traced, on-demand, inside the reconstructed scene. The color information retrieved from these lumigraphs is then used to estimate a linear combination of basis BRDFs for a grid of points in the surface area, defining thus its SVBRDF. This paper details the proposed method and presents the results obtained using real and synthetic settings. It shows that considering the whole environment as a source of light is a viable approach to obtain reliable results and to enable more flexible acquisition setups.

Quality assessment of potatoes necessitates a computer aided interpretation technique that uses a nonhazardous and noncontact, Non-Destructive Testing technique for image acquisition and a feature extraction technique for extracting and representing the features and a classifier for performing classification/grading of potatoes. In this paper, InfraRed Thermography is used for image acquisition and Convolutional Neural Networks (CNNs) are used for the classification of potatoes. In this work, thermograms of Normal, Fungus affected potatoes, potatoes with Holes and worst affected potatoes are used. Feasibility of VGGNET, RESNET and modified ALEXNET for classification of potatoes is studied. Performance is measured in terms of sensitivity and accuracy. It is found that RESNET18 outperformed all the other networks in terms of accuracy and it is the only network that classified all the affected potatoes as affected potatoes, i.e. the affected potatoes are not wrongly identified as normal.

Palmprints are a recognizable biometric feature that contains multimodal information. Contact-based high-resolution palmprints are the most commonly used image in automatic palmprint identification systems with high-level security requirements because of its high matching accuracy and robustness. Currently, the traditional contact-based high-resolution palmprint acquisition system only captures the total internal reflection (TIR) image, which is similar to fingerprint acquisition systems. However, it is worth noting that the palm is much larger than the finger, and different regions of the palm cannot be easily pressed to the same plane. We observed that the images obtained by these existing contact-based high-resolution palmprint image acquisition systems often have poor integrity and sharpness, especially in the center of the palm. Considering this problem, we proposed an improved contact-based high-resolution palmprint image acquisition system. In addition to the TIR image, it can also simultaneously acquire a diffuse reflection (DR) image. A high-resolution DR image has low contrast and is thus not suitable for direct identification. However, the high integrity and sharing of structural information with the TIR image make it possible to complement the TIR image. Our strategy is to transfer the DR image to the TIR domain through a convolution neural network (CNN) and then fuse it with the corresponding TIR image to obtain a higher quality TIR domain image. The partial-to-“full” identification experimental results show that the rank-1 accuracy of the fused TIR (FTIR) image is improved by approximately 23% compared with the traditional TIR image, which proves the effectiveness of our strategy.

High magnification optical cameras, such as microscopes or macro-photography, cannot capture an object that is totally in focus. In this case, image acquisition is done by capturing the object/scen...

This paper presents field experiments to test a human-portable towed remotely operated vehicle developed for shallow seabed image data acquisition at high speeds and over a wide range. Through analysis and fundamental experiments, we designed and developed the towed ROV with a multi-camera unit. The portable vehicle can be operated even from small boats. Therefore, it offers an alternative survey method to a Manta tow survey guided by divers. We also conducted field experiments to verify the effectiveness of its motion performance and image acquisition performance. The multi-camera unit mounted on the towed ROV recorded seabed images at several speeds of 0.67–1.52 m/s. After the experiments, we built several 3D models of the seabed offline using experimentally obtained images. From those experimentally obtained results, we conclude that this towed ROV with the multi-camera unit is promising for use in wide area observations and periodic monitoring of shallow seabed areas.

In vivo reflectance confocal microscopy (RCM) enables clinicians to examine lesions’ morphological and cytological information in epidermal and dermal layers while reducing the need for biopsies. As RCM is being adopted more widely, the workflow is expanding from real-time diagnosis at the bedside to include a capture, store, and forward model with image interpretation and diagnosis occurring offsite, similar to radiology. As the patient may no longer be present at the time of image interpretation, quality assurance is key during image acquisition. Herein, we introduce a quality assurance process by means of automatically quantifying diagnostically uninformative areas within the lesional area by using RCM and coregistered dermoscopy images together. We trained and validated a pixel-level segmentation model on 117 RCM mosaics collected by international collaborators. The model delineates diagnostically uninformative areas with 82% sensitivity and 93% specificity. We further tested the model on a separate set of 372 coregistered RCM-dermoscopic image pairs and illustrate how the results of the RCM-only model can be improved via a multimodal (RCM + dermoscopy) approach, which can help quantify the uninformative regions within the lesional area. Our data suggest that machine learning–based automatic quantification offers a feasible objective quality control measure for RCM imaging.

Sidestream dark field (SDF) imaging enables direct visualisation of the microvasculature from which quantification of key variables is possible. The new MicroScan USB3 (MS-U) video-microscope is a hand-held SDF device that has undergone significant technical upgrades from its predecessor, the MicroScan Analogue (MS-A). The MS-U claims superior quality of sublingual microcirculatory image acquisition over the MS-A, however, this has yet to be robustly confirmed. In this manuscript, we therefore compare the quality of image acquisition between these two devices. The microcirculation of healthy volunteers was visualised to generate thirty video images for each device. Two independent raters, blinded to the device type, graded the quality of the images according to the six different traits in the Microcirculation Image Quality Score (MIQS) system. Chi-squared tests and Kappa statistics were used to compare not only the distribution of scores between the devices, but also agreement between raters. MS-U showed superior image quality over MS-A in three of out six MIQS traits; MS-U had significantly more optimal images by illumination (MS-U 95% optimal images, MS-A 70% optimal images (p-value 0.003)), by focus (MS-U 70% optimal images, MS-A 35% optimal images (p-value 0.002)) and by pressure (MS-U 72.5% optimal images, MS-A 47.5% optimal images (p-value 0.02)). For each trait, there was at least 85% agreement between the raters, and all the scores for each trait were independent of the rater (all p-values > 0.05). These results show that the new MS-U provides a superior quality of sublingual microcirculatory image acquisition when compared to old MS-A

Background Quantitative and qualitative assessment of visual and morphological traits of seed is slow and imprecise with potential for bias to be introduced when gathered with handheld tools. Colour, size and shape traits can be acquired from properly calibrated seed images. New automated tools were requested to improve data acquisition efficacy with an emphasis on developing research workflows. Results A portable imaging system (BELT) supported by image acquisition and analysis software (phenoSEED) was created for small-seed optical analysis. Lentil (Lens culinaris L.) phenotyping was used as the primary test case. Seeds were loaded into the system and all seeds in a sample were automatically individually imaged to acquire top and side views as they passed through an imaging chamber. A Python analysis script applied a colour calibration and extracted quantifiable traits of seed colour, size and shape. Extraction of lentil seed coat patterning was implemented to further describe the seed coat. The use of this device was forecasted to eliminate operator biases, increase the rate of acquisition of traits, and capture qualitative information about traits that have been historically analyzed by eye. Conclusions Increased precision and higher rates of data acquisition compared to traditional techniques will help to extract larger datasets and explore more research questions. The system presented is available as an open-source project for academic and non-commercial use.

Image segmentation is essential in object-based image analysis. Numerous image segmentation algorithms have been proposed and widely applied to process remote sensing images, but most of them are designed to deal with single scenes. As the volume of images grows rapidly, handling images with single machines is becoming increasingly difficult, and the size of a composite image can be larger than the CPU memory of a single computer. To address this problem, a distributed image segmentation strategy is proposed in this paper. The two main steps of the proposed strategy are as follows. First, a prepared massive image is loaded and then decomposed into sub-images that are distributed across multiple computers; algorithms are then used in parallel to segment each sub-image into a large number of initial objects. Secondly, the proposed object resegmentation method is applied to the initial boundary objects in each sub-image in order to merge these objects. The sub-images are then ingested from the different computers in order to obtain the final segmentation image. Two classical segmentation algorithms are employed to test the proposed strategy in eight different study areas that include urban area, suburban zone and agricultural landscape. Both the intersection over union and the F-measure metrics show that the proposed strategy can help to solve the problem of the data volume being too large to fit on a single machine, and that it also performs better than comparative strategies. The proposed strategy not only has the ability to segment very large images, but also accelerates the segmentation and segmentation-based applications so that they can match the image acquisition rate.

In this review article, the current and future impact of artificial intelligence (AI) technologies on diagnostic imaging is discussed, with a focus on cardio-thoracic applications. The processing of imaging data is described at 4 levels of increasing complexity and wider implications. At the examination level, AI aims at improving, simplifying, and standardizing image acquisition and processing. Systems for AI-driven automatic patient iso-centering before a computed tomography (CT) scan, patient-specific adaptation of image acquisition parameters, and creation of optimized and standardized visualizations, for example, automatic rib-unfolding, are discussed. At the reading and reporting levels, AI focuses on automatic detection and characterization of features and on automatic measurements in the images. A recently introduced AI system for chest CT imaging is presented that reports specific findings such as nodules, low-attenuation parenchyma, and coronary calcifications, including automatic measurements of, for example, aortic diameters. At the prediction and prescription levels, AI focuses on risk prediction and stratification, as opposed to merely detecting, measuring, and quantifying images. An AI-based approach for individualizing radiation dose in lung stereotactic body radiotherapy is discussed. The digital twin is presented as a concept of individualized computational modeling of human physiology, with AI-based CT-fractional flow reserve modeling as a first example. Finally, at the cohort and population analysis levels, the focus of AI shifts from clinical decision-making to operational decisions.

Seven commercial Chinese chrysanthemum tea products were classified by computer vision combined with machine learning algorithms. Without the need of building any specific hardware, the image acquisition was achieved in two computer vision approaches. In the first approach, a series of multivariate classification models were built after morphological feature extraction of the image. The best prediction accuracies when classifying flowering stages and tea types were respectively 90% and 63%. In comparison, the deep neural network was applied directly on the raw image, yielded 96% and 89% correct identifications when classifying flowering stage and tea type, respectively. The model can be applied for rapid and automatic quality determination of teas and other related foods. The result indicated that computer vision, especially when combined with deep learning or other machine learning techniques can be a convenient and versatile method in the evaluation of food quality., This manuscript reported the multivariate classification of seven commercial Chinese chrysanthemum tea products by computer vision combined with machine learning algorithms. The result indicated that computer vision, especially when combined with deep learning or other machine learning techniques can be a convenient and versatile method in the evaluation of food quality.

Objectives Evaluation of tongue colour is an important approach for assessment of human health in traditional East Asian medicine, which originated in ancient China. However, tongue colour analyses are unreliable due to poor quantification and reproducibility. Given these limitations, the utility of this technique as a clinical index has not been demonstrated. Here, we assessed whether tongue colour evaluations, as performed in Kampo (traditional Japanese) medicine, via an automated image acquisition system were associated with particular indices of patient health. Methods We obtained high colour reproduction tongue images from 816 Japanese subjects and applied a machine-learning analysis approach to the quantification of tongue body colour in them. Data on sex, age, systolic blood pressure, diastolic blood pressure, pulse rate (PR), body temperature, body mass index (BMI), past medical history (PMH), and blood examination results (haemoglobin, Hb; creatinine) were also collected. Results Tongue body colour was influenced by multiple clinical indices, per simple correlation analyses. In particular, sex, age, BMI, PR, and Hb were correlated with tongue body colour. No significant correlation with PMH was found. Conclusions Tongue body colour, as quantified with a machine learning approach, was found to be a clinical predictor of changes in several clinical indices, particularly those related to obesity and haemoglobin concentrations.

Image processing is mostly used for exploring image behaviour. There are several steps in image processing. Image acquisition, pre-processing, feature extraction, and classification are the processes used for the detection of human movement based on high-level feature extraction (HLFE), in which HLFE was used for feature extraction in this paper. This study proposed the use of background subtraction and frame difference. This research was conducted to analyse the difference of background subtraction and frame difference methods based on movement of human. Movement of human detected by using feature extraction were centroid image technique used. Furthermore, support vector machine (SVM) was used for classification.

With the rapid development and advancement of society, information technology represented by biometric recognition has developed rapidly. Palmprint images have a larger area, more texture details, rich texture features and distinguishable information, and palmprint recognition is opposite to the palm. The pattern image acquisition equipment and resolution requirements are not high, which is suitable for palmprint image acquisition and analysis under the framework of the Internet of Things (IoT), and then realize the design of recognition technology and system based on palmprint texture characteristics. The palmprint texture feature extraction and analysis algorithm is deployed on the cloud server, and the analysis result is fed back to the on-site collection device to realize the collection and analysis of palmprint images.

It is possible to identify the existing license plate (LPR) pictures or videos depending on the algorithms used and in general it is possible that the algorithms consist of three steps in terms of processing are 1) extracting the part of the license plate; 2) manufacture of the license plate section. and 3) recognition of each character. The process is difficult due to the diversity of the panels' formats, as well as the lack of sufficient and uniform lighting during image acquisition, so it will be constrained under conditions such as fixed lighting, speed, specific roads, and other conditions. Several LPR technologies were developed in images and videos. This paper is a review of the authors work in this field and extracting the best results that the researchers reached to evaluate the LPR algorithm.