Automated evaluation of optical microscopy images of liquid jets, commonly used for sample delivery at X-ray free-electron lasers (XFELs), enables real-time tracking of the jet position and liquid jet hit rates, defin...
详细信息
Automated evaluation of optical microscopy images of liquid jets, commonly used for sample delivery at X-ray free-electron lasers (XFELs), enables real-time tracking of the jet position and liquid jet hit rates, defined here as the proportion of XFEL pulses intersecting with the liquid jet. This method utilizes machinevision for preprocessing, feature extraction, segmentation and jet detection as well as tracking to extract key physical characteristics (such as the jet angle) from optical microscopy images captured during experiments. To determine the effectiveness of these tools in monitoring jet stability and enhancing sample delivery efficiency, we conducted XFEL experiments with various sample compositions (pure water, buffer and buffer with crystals), nozzle designs and jetting conditions. We integrated our real-time analysis algorithm into the Karabo control system at the European XFEL. The results indicate that the algorithm performs well in monitoring the jet angle and provides a quantitative characterization of liquid jet stability through optical image analysis conducted during experiments.
Crops and weeds are involved in a continuous competition for equal resources, which may result in a potential decrease in crop yields by up to 31% and an increase in the costs of agricultural inputs by up to 22% of cu...
详细信息
Crops and weeds are involved in a continuous competition for equal resources, which may result in a potential decrease in crop yields by up to 31% and an increase in the costs of agricultural inputs by up to 22% of cultivation. Weeds further impact crop production, and their detection is crucial for effective crop management. In this research, we targeted common weeds of cotton field, specifically i) Digitaria sanguinalis (L.) Scop, ii) Amaranthus retroflexus L., iii) Acalypha australis, L., iv) Cephalanoplos segetum, and v) Chenopodium album L. Additionally, image processing techniques such as grayscale conversion, binarization, and Gaussian and morphological filters were also utilized. These methods are based on machinevision and facilitate rapid and straightforward weed detection by segmenting, scrutinizing, and comparing input images. The plant height and area were obtained during cotton planting within 32 days and fitted to develop the growth law concerning planting days for achieving the function of distinguishing cotton from weeds. We conducted recognition experiments by dividing images into four quadrants and categorizing weeds as either inter-row or intra-row. Meanwhile, the inter-row planting information was used to identify weeds, and the leaf pixel area and circularity were used as the identification methods for intra-row weeds, which reduced the algorithm's running time and improved real-time performance. The experimental results indicated that the inter-row weed recognition rate was 89.4%, with an average processing time of 102ms. Whereas in the case of intra-row weeds, the recognition rate was measured at 84.6%, and the overall recognition rate for cotton was 85.0%, with a mean time consumption of 437ms. Furthermore, the present research underscores recent advancements such as machinevision and high-resolution imaging, which have significantly improved the accuracy of automated weed identification in cotton fields while acknowledging ongoing challen
In this paper, we propose a concept for AI and machinevision for the observation and assessment of induced pluripotent stem (iPS) cell culture based on a differentiation of the two major monitoring aspects of cell ex...
详细信息
ISBN:
(纸本)9781665493130
In this paper, we propose a concept for AI and machinevision for the observation and assessment of induced pluripotent stem (iPS) cell culture based on a differentiation of the two major monitoring aspects of cell expansion - cell health and cell density. The proposed concept is part of a more holistic approach to automating cell cultures in smaller laboratories. The concept embedded the broader automation of basic cell culture procedures using iPS cells and expansion protocols, as well as implementing mechanisms and a multipurpose gripper for discarding old and refilling fresh culture media and handling individual vessels. Based on phase contrast microscopy imaging our approach involves utilizing machine vision algorithms to calculate cell density and decide whether to split the culture or not. We implemented an image algorithm to achieve this and used threshold values derived from the working experiences with iPS cells. We concluded that our approach's achieved accuracy is sufficient to automate this task of cell assessment. The process is modular so that the proposed algorithms can be implemented with manually taken images or fully automated imaging. A trained image-based AI can be used to obtain a decision if the shown phase contrast image only shows healthy cells. If there are any not trained formations, the cells will be discarded. Future work includes completing the set of images for training the AI to recognize any deviation i.e. contamination, premature differentiation, or cell death for the health aspect of cell expansion monitoring. We aim to streamline and automate basic iPS cell culture procedures and make them more accessible to smaller laboratories. The proposed algorithms for cell health and available growing space assessment seem promising and may help with decision-making and ensure the health and growth of the iPS cells.
Due to the complex structure of power cables, which are often laid underground, if a fault occurs and cannot be located and repaired in a timely manner, it would bring huge economic losses. In order to ensure the safe...
详细信息
The inspection of defects is an important task in many industrial sectors: from metals to plastics, passing through glass and other materials, these products need to satisfy some aesthetical and quality requirements. ...
详细信息
ISBN:
(数字)9781510627925
ISBN:
(纸本)9781510627925
The inspection of defects is an important task in many industrial sectors: from metals to plastics, passing through glass and other materials, these products need to satisfy some aesthetical and quality requirements. Flaws can arise in many different forms: spot of different color, crack, incompleteness, excess and/or lack of material are just some examples of defects deriving from the industrial manufacturing process, which can lead to discard the component or the piece examined. These defects are recognizable by the human eye, but some issues like fatigue, illness of the operator and incorrect lighting of the samples can be tough obstacles in obtaining the right selection of the pieces. To detect faulty pieces and in order to avoid wasting compliant pieces instead, a computer based visual inspection system has been designed and implemented. As benchmark samples we adopt the outer lenses of automotive rear lamps. The surface of an outer lens needs an extreme precision manufacturing procedure and the absence of defects is essential for the quality of the final product. The aim of the work involves the ideation and commissioning of a setup to extract and analyze information about the flaws present in an outer lens, exploiting different image processing techniques depending on the nature of the defects.
The thread measuring method possesses an efficient and easy to install with image measurement of axial section but results in thread profile distortion. Through the geometric analysis of the axial section projection i...
详细信息
The thread measuring method possesses an efficient and easy to install with image measurement of axial section but results in thread profile distortion. Through the geometric analysis of the axial section projection image of the screw thread, it is confirmed that phenomenon of thread profile distortion always exist in measuring screw thread by projection image, meanwhile, the requirements of depth of focus is put forward with the method, calculation formula of thread profile distortion is deduced, and the corresponding compensation algorithm is given;through the simulation, analysis of coarse thread, it found the distortion has influence in thread angle and pitch diameter significantly, reaching 2.3% and 3.36% respectively, the unilateral shaded proportion of the thread profile in the root region is up to 7.12%, but the root diameter itself remains unaffected mostly. Take metric coarse thread gauge as measurement object, and compared the results with tool microscope, experiments show that the compensation effect is more than 85% in pitch diameter, more than 70% in thread angle.
This paper addresses a problem of video-based face recognition. We propose a new neural network model that uses an input set of facial images of a person to produce a compact, fixed-dimension descriptor. Our model is ...
详细信息
This paper addresses a problem of video-based face recognition. We propose a new neural network model that uses an input set of facial images of a person to produce a compact, fixed-dimension descriptor. Our model is composed of two modules. The feature embedding module maps each image onto a feature vector, while the face quality assessment module estimates the utility of each facial image. These feature vectors are weighted based on their utility estimations, resulting in the image set feature representation. During visual analysis we found that our model learns to use more information from high-quality face images and less information from blurred or occluded images. The experiments on YouTube Faces and Janus Benchmark A (IJB-A) datasets show that the proposed feature aggregation method based on face quality assessment consistently outperforms naive aggregation methods.
A vision-based measurement system to quantify the yarn density of woven fabrics during production is presented. As an extension to an earlier developed fabric flaw detection system, the proposed framework consists of ...
详细信息
A vision-based measurement system to quantify the yarn density of woven fabrics during production is presented. As an extension to an earlier developed fabric flaw detection system, the proposed framework consists of a combination of basic and custom-made image-processing techniques that allow to precisely track single wefts and warps within fabric images-in real-time. Several adaptations facilitate the measurement of density changes for plain, satin, and twill weaves. In this paper, the algorithmic framework has been evaluated in several comprehensive on-line experiments on a real-world air-jet loom and is additionally compared with three alternative methods for fabric density measurement. It proved to be precise, robust, and applicable for industrial use as it overcomes many of the existing shortcomings of current methods.
This paper addresses a problem of false positive detection filtering in surveillance video streams. We propose two methods. The first one is based on automatic hard negative mining from a video stream, which is then u...
详细信息
This paper addresses a problem of false positive detection filtering in surveillance video streams. We propose two methods. The first one is based on automatic hard negative mining from a video stream, which is then used for fine-tuning of the baseline detector. The second one is the detector output filtering by analyzing the frequency of detection of visually similar samples. We demonstrate the proposed methods on cascade-based detectors, but they can be applied to any detector that can be trained in a reasonable amount of time. Experimental results show that the proposed methods improve both the precision and recall rate, as well as reducing the computational time by 47%.
Mathematical morphology is inheriently suitable for range image processing because it can deal with the shape of a function in a natural and intuitive way. In this paper, a new approach to the extraction of the corner...
详细信息
Mathematical morphology is inheriently suitable for range image processing because it can deal with the shape of a function in a natural and intuitive way. In this paper, a new approach to the extraction of the corner-edge-surface structure from 3D range images is proposed. Morphological operations are utilized for segmenting range images into smooth surface regions and high-variation surface regions, where the high-variation surface regions are further segmented into regions of edge type and regions of corner type. A new 3D feature, HV-skeleton, can be extracted for each high-variation surface region. The HV-skeletons can be thought of as the skeletons of high-variation surface regions and are useful for feature matching. The 3D features extracted by our approach are invariant to 3D translations and rotations, and can be utilized for higher level vision tasks such as registration and recognition. Experimental results show that the new 3D feature extraction method works well for both simple geometric objects and complex shaped objects such as human faces.
暂无评论