(Semi-)automatic recognition of microorganisms in water

An automatic microscope image acquisition, evaluation, and recognition system was developed for the analysis of Utermöhl plankton chambers in terms of taxonomic algae recognition. The system called PLASA (Plankton Structure Analysis) comprises (1) fully automatic archiving (optical fixation) of aqueous specimens as digital bright field and fluorescence images, (2) phytoplankton analysis and recognition, and (3) training facilities for new taxa. It enables characterization of aqueous specimens by their populations. The system is described in detail with emphasis on image analytical aspects. Plankton chambers are scanned by sizable grids, divers objective(s), and up to four fluorescence spectral bands. Acquisition positions are focused and digitized by a TV camera and archived on disk. The image data sets are evaluated by a large set of quantitative features. Automatic classifications for a number of organisms are developed and embedded in the program. Interactive programs for the design of training sets were additionally implemented. A long-term sampling period of 23 weeks from two ponds at two different locations each was performed to generate a reliable data set for training and testing purposes. These data were used to present this system's results for phytoplankton structure characterization. PLASA represents an automatic system, comprising all steps from specimen processing to algae identification up to species level and quantification.

In the age of ever-increasing "-omics" studies, the accurate and statistically robust determination of microbial cell numbers within often-complex samples remains a key task in microbial ecology. Microscopic quantification is still the only method to enumerate specific subgroups of microbial clades within complex communities by, for example, fluorescence in situ hybridization (FISH). In this study, we improved an existing automatic image acquisition and cell enumeration system and adapted it for usage at high seas on board an oceanographic research ship. The system was evaluated by testing settings such as minimal pixel area and image exposure times ashore under stable laboratory conditions before being brought on board and tested under various wind and wave conditions. The system was robust enough to produce high-quality images even with ship heaves of up to 3 m and pitch and roll angles of up to 6.3°. On board the research ship, on average, 25% of the images acquired from plankton samples on filter membranes could be used for cell enumeration. Automated enumeration was highly correlated with manual counts (r(2) > 0.9). Even the smallest of microbial cells in the open ocean, members of the alphaproteobacterial SAR11 clade, could be confidently detected and enumerated. The automated image acquisition and cell enumeration system developed here enables an accurate and reproducible determination of microbial cell counts in planktonic samples and allows insight into the abundance and distribution of specific microorganisms already on board within a few hours.IMPORTANCE In this research article, we report on a new system and software pipeline, which allows for an easy and quick image acquisition and the subsequent enumeration of cells in the acquired images. We put this pipeline through vigorous testing and compared it to manual microscopy counts of microbial cells on membrane filters. Furthermore, we tested this system at sea on board a marine research vessel and counted bacteria on board within a few hours after the retrieval of water samples. The imaging and counting system described here has been successfully applied to a number of laboratory-based studies and allowed the quantification of thousands of samples and FISH preparations (see, e.g., H. Teeling, B. M. Fuchs, D. Becher, C. Klockow, A. Gardebrecht, C. M. Bennke, M. Kassabgy, S. Huang, A. J. Mann, J. Waldmann, M. Weber, A. Klindworth, A. Otto, J. Lange, J. Bernhardt, C. Reinsch, M. Hecker, J. Peplies, F. D. Bockelmann, U. Callies, G. Gerdts, A. Wichels, K. H. Wiltshire, F. O. Glöckner, T. Schweder, and R. Amann, Science 336:608-611, 2012, http://dx.doi.org/10.1126/science.1218344). We adjusted the standard image acquisition software to withstand ship movements. This system will allow for more targeted sampling of the microbial community, leading to a better understanding of the role of microorganisms in the global oceans.

In liver stereotactic body radiotherapy (SBRT) using fiducial markers, the accuracy of automatic image recognition of fiducial markers is important, and the imaging dose cannot be neglected in image-guided radiotherapy. Optimal imaging parameters of fiducial markers were investigated for automatic image recognition and imaging dose. We investigated automatic recognition with fiducial markers of different shapes and sizes. In addition, the optimum imaging conditions were examined based on the automatic recognition when the presence or absence of a filter, focal spot size, and phantom thickness were altered using the fiducial markers with a high automatic recognition. The results for different shapes and sizes of fiducial markers showed that larger markers were recognized more automatically, whereas shorter markers were recognized in the correct position. By using the filter, we were able to reduce the imaging dose by one third or one half compared to the case without the filter. The results for the focal spot size showed that using a larger size resulted in higher automatic recognition accuracy than using a smaller size. For the relationship between the automatically recognized imaging conditions and the air kerma when the phantom thickness was altered, it was necessary to keep the tube current-time product constant and increase the tube voltage in order to avoid poor recognition accuracy. The parameters we proposed are effective in shortening the treatment time and reducing the imaging dose because they allow us to acquire images with low doses and high accuracy of automatic recognition.

Funding Acknowledgements Type of funding sources: Public grant(s) – EU funding. Main funding source(s): This publication is part of a project that has received funding from the European Union’s Horizon 2020 research and innovation program under grant agreement No 954783. Background Stereotactic arrhythmia radio-ablation (STAR) is a promising strategy for the non-invasive treatment of ventricular arrhythmias (VA). The management of the cardio-respiratory motion of the target and the reduction of the uncertainties related to patients’ positioning are two of the main challenges that STAR has to overcome. A prototype of a system was developed that can automatically acquire and interpret echocardiographic images using an artificial intelligence (AI) algorithm to calculate cardiac displacement in real-time. Purpose To evaluate the feasibility of this automatic acquisition system in patients with a history of VA. Methods We conducted a single center, feasibility study enrolling consecutive patients with a history of VA. Echocardiographic images were automatically acquired from the parasternal and apical views with a dedicated probe. The system was designed to hold the probe fixed to the chest in the supine position during both free-breathing and short expiratory breath-hold sequences, to simulate STAR treatment. The primary endpoint was the percentage of patients reaching a score ≥2 in a multi-parametric assessment evaluating the quality of automatically acquired images in terms of: A - allowing a proper identification of the cardiac cycle phase using the AI algorithm, B - allowing a correct measurement of the heart displacement by the AI algorithm, C - ability to visually distinguish typical cardiac structures, and D - stability of the image throughout the respiration cycle. Moreover, we investigated the potential impact of clinical and demographic characteristics on achieving the primary outcome. Results From May to September 2021, we enrolled 24 patients (63±14 years, 21% females). All of them had a history of VA and 21 (88%) had an ICD. Eight patients (33%) had coronary artery disease, 12 (50%) had non-ischemic cardiomyopathy, and 3 had idiopathic VA. Parasternal as well as apical images were obtained from all patients except from one, in whom parasternal view could not be collected due to the patient’s inability to maintain the supine position. The primary outcome was achieved in 23 patients (96%) for the apical view, in 20 patients (87%) for the parasternal view, and in all patients in at least one of the two views. The images quality was maximal (i.e. score=4) in at least one of the two windows in 19 patients (79%). Atrial fibrillation arrhythmia was the only clinical characteristics associated with a poor score outcome in both imaging windows (apical p=0,022, parasternal p=0,014). Conclusion An automatic ultrasonographic image acquisition system associated with an AI algorithm is feasible for real-time monitoring of cardiac motion in patients with a history of VA. The possibility of real-time, non-invasive monitoring of cardiac position would lead to a significant improvement in the quality and safety of STAR treatment, particularly in case of treatment with heavy particles such as protons and carbon ions.

In recent years, with the popularity of the Internet, digital media has become an indispensable part of people's life. Image acquisition equipment is widely used in daily life and work. However, due to technical problems, it can not meet the performance requirements of users. At the same time, a large number of data and information are difficult to be effectively processed, which also reduces the image quality, reduces the storage space and increases the image distortion. Therefore, it is necessary to develop new algorithms to improve these defects. Therefore, this paper designs and develops an automatic digital media image recognition system based on mean clustering algorithm. Firstly, this paper describes the concept and characteristics of digital media. Then, this paper also studies the application of mean clustering algorithm. Based on this, an automatic digital media image recognition system is designed, and the performance of the system is tested by simulation. Finally, the test results show that the algorithm can accurately identify media images. In general, the performance of the digital media image automatic recognition system based on mean clustering algorithm can meet the needs of target users.

Visibility detection has great research and practical application value in the fields of weather forecasting and autonomous driving. At the same time, the existing visibility detection methods have some problems, such as the high price of hardware equipment, the need to set specific markers, and the large amount of calculation. In this paper, a visibility detection algorithm based on automatic image recognition is proposed, which calculates the light intensity of the target area of the image through a series of image processing, calculates the corresponding transmittance through the dark channel prior algorithm, and then inverts the atmospheric visibility according to the Koschmieder atmospheric visibility measurement principle. The corresponding fog concentration coefficient under different visibility conditions was measured by experiments, and the visibility value calculated based on the image and the visibility value of the CJY-1G forward visibility scattering instrument were compared, and the error percentage before the two was calculated, so as to verify that the visibility value detected by the method used in this chapter can fully meet the specified error range and meet the application requirements of visibility observation.

In modern clinical medicine, the important information of red blood cells, such as shape and number, is applied to detect blood diseases. However, the automatic recognition problem of single cells and adherent cells always exists in a densely distributed medical scene, which is difficult to solve for both the traditional detection algorithms with lower recognition rates and the conventional networks with weaker feature extraction capabilities. In this paper, an automatic recognition method of adherent blood cells with dense distribution is proposed. Based on the Faster R-CNN, the balanced feature pyramid structure, deformable convolution network, and efficient pyramid split attention mechanism are adopted to automatically recognize the blood cells under the conditions of dense distribution, extrusion deformation, adhesion and overlap. In addition, the Align algorithm for region of interest also contributes to improving the accuracy of recognition results. The experimental results show that the mean average precision of cell detection is 0.895, which is 24.5% higher than that of the original network model. Compared with the one-stage mainstream networks, the presented network has a stronger feature extraction capability. The proposed method is suitable for identifying single cells and adherent cells with dense distribution in the actual medical scene.

Spiders are the most abundant predatory natural enemies in terrestrial ecosystems. As an important natural enemy of many agricultural and forestry pests, spiders play a very significant role in the biological control of pests. In order to make rational use of spider resources, it is necessary to observe and study the population characteristics of it. Direct observation method is time-consuming and laborious. If we can take the videos or images of spiders by surveillance cameras, and then use computer vision technology to identify and classify automatically, the efficiency of image data acquisition and pest biological control will be greatly improved. Motivated by this, we studied the classification and recognition of spider images in natural background obtained by common surveillance equipment. However, the images of some species of spiders in natural background are difficult to be collected, and the inadequate clarity and contrast of the subjects in images will also affect the recognition accuracy. So, firstly, histogram equalization was used to enhance the contrast of the image; the dataset of spider images was expanded by image rotation, reflection, flipping, zooming, translating and increasing the pixel noise appropriately, and so on; the contour detection was carried out for assistant recognition. Secondly, taken the deep convolutional neural networks (CNN) as our basic framework, two automatic recognition models of spider images, that is the 8-layer deep CNN model and the transfer learning model based on Inception-v3, were constructed. After that, the two models were trained, evaluated and compared under a dataset with 4478 pre-processed images. The experimental results show that the first model has a limited effect on image feature extraction of spiders in natural background, while the second model based on transfer learning can achieve better recognition accuracy when combining image contour features as an auxiliary input. In the second model, the accuracy of training set and testing set can reach more than 90%, and the recognition speed can be controlled within 1 second, which meets the practical requirements of automatic classification and recognition of spider images in natural background.

With the rapid growth of technology and the proliferation of data in this digital age, current image and audio applications require greater resolution, higher data transmission rates and better data compression techniques to meet the ever increasing demands placed on them. The research presented here investigates the impact of data compression in the automatic recognition of handwritten digit images and spoken digit audio. A Haar wavelet transform (HWT) is used to compress the original image and audio data, which is input to an artificial neural network (ANN) where the automatic digit recognition is performed. The HWT generates a signature, or fingerprint, for the data by removing redundant data using a cut-off function, a number of which are investigated. This reduced data signature enables the ANN-based recogniser to be simplified and computationally more efficient. Experimental results show that for handwritten digit images, the recognition accuracy is 94.3% with compression ratios of 80%; for spoken audio digits, the recognition accuracy is 98.8% with compression ratios of 82%.

gamma-H2AX is emerging as an important marker of ionizing radiation-induced double-strand breaks. Development of a significantly automated method to quantify gamma-H2AX would have broad application in assessing physiological responses to radiation exposure. PC-3 and DU145 prostate cancer cells grown on glass cover slips and 96-well plates were irradiated and assessed for gamma-H2AX focus formation by immunofluorescence analysis. The gamma-H2AX immunofluorescence staining was performed either manually or by using a preprogrammed automated robotic liquid handling system. A computer-controlled charge-coupled device camera acquired images serially throughout the thickness of each cell. Image analysis was performed manually and/or with automated image segmentation software. A robust relationship between radiation dose and gamma-H2AX focus numbers was demonstrated with both manual and automated image analysis methods, with excellent agreement observed between the two techniques. The r(2) correlation coefficients and Z factors exceeded 0.9 and 0.5, respectively, when gamma-H2AX focus formation was correlated with radiation dose using the automated technique. Inhibition of gamma-H2AX foci by drugs readily detected with this assay. Robotic specimen preparation with automated image acquisition and analysis can be used to quantify gamma-H2AX foci in irradiated cells, and the results agree well those obtained by manual counts. These data suggest that this assay has an excellent signal-to-noise ratio and is suitable for high-throughput applications.

With the continuous improvement of intelligent operation and inspection requirements of transmission lines in power system, the popularization of online monitoring of transmission lines and uav inspection, a large number of transmission line defect image data are generated. However, traditional identification technology has low efficiency and inaccurate detection in the identification of transmission line defects. Based on transmission line defect source data, automatic transmission line defect image recognition technology based on artificial intelligence is studied. The transmission line defect identification technology is optimized and improved through automatic video image defect identification and defect sample annotation technology of transmission line components. The field scene verification proves that, compared with the traditional identification technology, the proposed adaptive algorithm for the selected area of transmission line defects can effectively improve the identification efficiency of transmission line defects, and the identification accuracy is close to 100%, which provides a guarantee for the stable operation of the power transmission system in the power industry.

This paper presents the results of experiments on the automatic recognition of few-pixel images in comparison with the data of visual recognition. The automatic recognition of few-pixel images is accomplished by means of an adaptive algorithm based on a compact description of the objects of an alphabet in the form of functional dependences of the mean values of the shape attributes and their rms deviations on the size of the image of an object and by using a fuzzy deciding rule with computation of confidence measures to hypotheses concerning the relation of the object to be recognized, applied to one of the objects of the alphabet. It is shown that applying an interpolation procedure to few-pixel images significantly increases the operating efficiency of the adaptive algorithm.

Chest X-ray is the commonly used method to diagnose pneumonia. How to correctly interpret the image information is always the main challenge faced by doctors. Convolution Neural Network (CNN) is a popular deep learning algorithm with excellent image recognition performance, and has been used widely in automatic recognition and diagnosis of medical images. This paper studies the classification of normal and pneumonia with more than 5000 chest X-ray images by employing three CNN models of VGG16, VGG19 and Inception_V3. The performances of each model for classification was evaluated and compared.

With the application of artificial intelligence technology, automatic recognition of brain image features has reached the stage of application. At present, the results of automatic recognition and labeling of brain images are displayed in two-dimensional form, which is not easy for doctors to observe intuitively. In addition, in the real medical environment, it is necessary to combine the virtual and the real, and to display some additional information in the real environment. Augmented reality can meet the above practical business needs. This paper studies the process of feature extraction and three-dimensional reconstruction of brain image diseases, designs the function module of augmented reality, explores the application of augmented reality in three-dimensional reconstruction, and realizes the three-dimensional reconstruction and visualization of brain image features by programming with Unity tool.

This paper presents a part of research study dealing with the initialization of an information system on radar automatic target recognition (ATR) chain. In this framework, we propose a set of helpful information at different levels of the ATR chain for decision making. The methodology used is the knowledge discovery process from data (KDD process).we focused in this paper on the data preparation step, more precisely Speckle denoising, edge detection and computation of features vector. We present the results of a several speckle filters that are largely used by ISAR imaging scientists (Lee, Frost, Kuan...). Fourier Descriptors have been used as features of the objects. Various similarity measures have been used and compared for target recognition.