An artificial neural network approach to identify fungal diseases of cucumber (Cucumis sativus L.) plants using digital image processing

Nowadays, artificial intelligence solutions such as digital image processing and artificial neural networks (ANN) have become important applicable techniques in phytomonitoring and plant health detection systems. In this research, an autonomous device was designed and developed for detecting two types of fungi (Pseudoperonospora cubensis, Sphaerotheca fuliginea) that infect the cucumber (Cucumis sativus L.) plant leaves. This device was able to recognise the fungal diseases of plants by detecting their symptoms on plant leaves (downy mildew and powdery mildew). For leaves of cucumber inoculated with different spores of the fungi, it was possible to estimate the amount of hour post inoculation (HPI) by extracting leaves’ image parameters. Device included a dark chamber, a CCD digital camera, a thermal camera, a light dependent resistor lightening module and a personal computer. The proposed programme for precise disease detection was based on an image processing algorithm and ANN. Three textural features and two thermal parameters from the obtained images were measured and normalised. Performance of ANN model was tested successfully for disease recognition and detecting HPI in images using back-propagation supervised learning method and inspection data. Such this machine vision system can be used in robotic intelligent systems to achieve a modern farmer’s assistant in agricultural crop fields.     

Computer-aided microtomography with true 3-D display in electron microscopy

A novel research system has been designed to permit three-dimensional (3-D) viewing of high resolution image data from transmission electron microscopy (TEM) and scanning electron microscopy (SEM). The system consists of front-end primary data acquisition devices, such as TEM and SEM machines, which are equipped with computer-controlled specimen tilt stages. The output from these machines is in analogue form, where a video camera attached to the TEM provides the sequential analogue image output while the SEM direct video output is utilized. A 10 MHz digitizer transforms the video image to a digital array of 512 X 512 pixel units of 8 bits deep-stored in a frame buffer. Digital images from multiple projections are reconstructed into 3-D image boxes in a dedicated computer. Attached to the computer is a powerful true 3-D display device which has hardware for graphic manipulations including tilt and rotate on any axis and for probing the image with a 3-D cursor. Data editing and automatic contouring functions are used to enhance areas of interest, and specialized software is available for measurement of numbers, distances, areas, and volumes. With proper archiving of reconstructed image sequences, a dynamic 3-D presentation is possible. The microtomography system is highly versatile and can process image data on-line or from remote sites from which data records would typically be transported on computer tape, video tape, or floppy disk.     

Automated Enumeration of Groups of Marine Picoplankton after Fluorescence In Situ Hybridization

We describe here an automated system for the counting of multiple samples of double-stained microbial cells on sections of membrane filters. The application integrates an epifluorescence microscope equipped with motorized z-axis drive, shutters, and filter wheels with a scanning stage, a digital camera, and image analysis software. The relative abundances of specific microbial taxa are quantified in samples of marine picoplankton, as detected by fluorescence in situ hybridization (FISH) and catalyzed reporter deposition. Pairs of microscopic images are automatically acquired from numerous positions at two wavelengths, and microbial cells with both general DNA and FISH staining are counted after object edge detection and signal-to-background ratio thresholding. Microscopic fields that are inappropriate for cell counting are automatically excluded prior to measurements. Two nested walk paths guide the device across a series of triangular preparations until a user-defined number of total cells has been analyzed per sample. A backup autofocusing routine at incident light allows automated refocusing between individual samples and can reestablish the focal plane after fatal focusing errors at epifluorescence illumination. The system was calibrated to produce relative abundances of FISH-stained cells in North Sea samples that were comparable to results obtained by manual evaluation. Up to 28 preparations could be analyzed within 4 h without operator interference. The device was subsequently applied for the counting of different microbial populations in incubation series of North Sea waters. Automated digital microscopy greatly facilitates the processing of numerous FISH-stained samples and might thus open new perspectives for bacterioplankton population ecology.     

Image analysis for monitoring the barley tempeh fermentation process

AIMS: To develop a fast, accurate, objective and nondestructive method for monitoring barley tempeh fermentation. METHODS AND RESULTS: Barley tempeh is a food made from pearled barley grains fermented with Rhizopus oligosporus. Rhizopus oligosporus growth is important for tempeh quality, but quantifying its growth is difficult and laborious. A system was developed for analysing digital images of fermentation stages using two image processing methods. The first employed statistical measures sensitive to image colour and surface structure, and these statistical measures were highly correlated (r=0.92, n=75, P<0.001) with ergosterol content of tempeh fermented with R. oligosporus and lactic acid bacteria (LAB). In the second method, an image-processing algorithm optimized to changes in images of final tempeh products was developed to measure number of visible barley grains. A threshold of 5 visible grains per Petri dish indicated complete tempeh fermentation. When images of tempeh cakes fermented with different inoculation levels of R. oligosporus were analysed the results from the two image processing methods were in good agreement. CONCLUSION: Image processing proved suitable for monitoring barley tempeh fermentation. The method avoids sampling, is nonintrusive, and only requires a digital camera with good resolution and image analysis software. SIGNIFICANCE AND IMPACT OF THE STUDY: The system provides a rapid visualization of tempeh product maturation and qualities during fermentation. Automated online monitoring of tempeh fermentation by coupling automated image acquisition with image processing software could be further developed for process control.     

A charge coupled device-based image cytophotometry system for quantitative histochemistry and cytochemistry

A rapid, semiautomated cytophotometry system for quantitative histochemistry and cytochemistry was constructed. The system consists of a Fairchild charge coupled device (CCD) image camera, a Zeiss Universal microscope, a Datacube analog to digital converter, and a digital Equipment Corporation LSI 11/23 computer operating under RT-11. Computer programs were written in FORTRAN and the MACRO assembly language for the acquisition of data from the CCD device. These data were then used for image segmentation, image display, and calculation of total optical density, perimeter, cell area, and several shape features. The reproducibility of measurement made with the CCD-based cytophotometry system was tested by repeated measurements. The coefficient of variation was estimated to be 1.7% for total optical density and 0.9% for cell area. The CCD-based cytophotometry system was further evaluated by comparing results with measurements made on the same cells with a scanning stage cytophotometer using the HIDACSYS computer programs. Correlation coefficients of 0.96 for total optical density and 0.91 for cell area were obtained between the two systems. We conclude that the high-speed, dimensional stability, small size, and linearity of the CCD-based cytophotometry system will make it useful for quantitative histochemistry and cytochemistry.     

Real time in situ microscopy for animal cell-concentration monitoring during high density culture in bioreactor

An in situ microscope (ISM) device is utilised in this study to monitor hybridoma cells concentration in a stirred bioreactor. It generates images by using pulsed illumination of the liquid broth synchronised with the camera frame generation to avoid blur from the cell's motion. An appropriate image processing isolates the sharp objects from the blurred ones that are far from the focal plane. As image processing involves several parameters, this paper focuses on the robustness of the results of the cells counting. This stage determines the applicability of the measuring device and has seldom been tackled in the presentations of ISM devices. Calibration is secondly performed for assessing the cell-concentration from the cell automated numeration provided by the ISM. Flow cytometry and hemacytometer chamber were used as reference analytical methods. These measures and the output of the image processing allow estimating a single calibration parameter: the reference volume per image equal to 1.08 x 10(-6) mL. In these conditions, the correlation coefficient between both reference and ISM data sets becomes equal to 0.99. A saturation of this system during an ultrasonic wave perfusion phase that deeply changes the culture conditions is observed and discussed. Principal component analysis (PCA) is used to undergo the robustness study and the ISM calibration step.     

Fully automated and fast image analysis of autoradiographs with a TAS-Leitz. Determination of size,Feulgen fluorescence and grain counts of individual nuclei and their evaluation by a simplified cluster analysis

Summary A fully automatic analysis system based on television image analysis was developed to measure simultaneously three parameters in individual nuclei of microscopic autoradiographs prepared from mouse jejunal crypt cell squashes and ascites tumor cell smears: size, Feulgen fluorescence and reflection from silver grains. A dark light camera with an image intensified silicon tube (RCA-ISIT), an automatic scanning stage and an autofocus device were fitted to a Leitz-TAS microscope. The camera permitted localization of Feulgen stained nuclei and measurement of area and light intensity by means of incident of light fluorescence in the red. After automatic changes of the Opak-illuminator silver grains were determined by means of polarized incident light reflected from the grains in the blue. A 25 x oil objective (aperture 0.75) yielded sufficient resolution for measurements. The nadir between the proportions of labeled and unlabeled nuclei was calculated from the data of one specimen on a PDP-computer using a new algorithm based on the minimal variance of the logarithm of reflected light per nucleus. Labeling indices determined by visual grain counting and by automatic analysis of the autoradiographs were well correlated (r=0.87 to 0.92). Visual grain counts/nucleus and reflected light/nucleus correlated well when individual nuclei were compared (r=0.92 to 0.97) or means of labeled nuclei of various specimens prepared during a 5 year period (r=0.90 to 0.93). Quenching of nuclear Feulgen fluorescence was minimal. The optimal labeling range is 30–100 grain counts/nucleus. The time interval between measurements of two specimens was 25 min for a squash of approximately 350 crypt cells within a 3 mm× 3 mm field, and 20 min for a meandering scan with 1,000 ascites tumor cells.     

In-situ microscopy: Online process monitoring of mammalian cell cultures

The in-situ microscope is a system developed to acquire images of mammalian cells directly inside a bioreactor (in-situ) duringa fermentation process. It requires only minimal operator intervention and it is well suited for either batch or long-termperfusion fermentation runs. The system fits into a 25 mm standard port and has a retractable housing, similar to the industry standard InTrac. Therefore, it can be cleaned and serviced without interruption of the process or risking contamination. A sampling zone inside the bioreactor encloses adefined volume of culture and an image sequence is taken. The height of the sampling zone is set by the control program and canbe adjusted during the cultivation to accommodate a wide range of change in cell density. The system has an infinity correctedoptical train and uses a progressive scan CCD camera to acquirehigh quality images. Process relevant information like cell density is extracted fromthe images by digital image processing software, currently in development for mammalian cells (CHO, BHK). The first version ofthe software will be able to estimate the cell density, cellsize distribution and to give information of the degree of aggregation (single and double cells, cell clusters).     

Microspectrofluorometry by digital image processing: measurement of cytoplasmic pH

An interface of our microspectrofluorometer with an image processing system performs microspectrofluorometric measurements in living cells by digital image processing. Fluorescence spectroscopic parameters can be measured by digital image processing directly from microscopic images of cells, and are automatically normalized for pathlength and accessible volume. Thus, an accurate cytoplasmic "map" of various spectroscopic parameters can be produced. The resting cytoplasmic pH of fibroblasts (3T3 cells) has been determined by measuring the ratio of fluorescein fluorescence exited by two successive wavelengths (489 and 452 nm). Fluorescein-labeled dextran microinjected into the cells is used as a pH indicator, since it is trapped in the cytoplasm but is excluded from the nucleus and other organelles. The average cytoplasmic pH is 6.83 (+/- 0.38). However, cytoplasmic pH exhibits a nonunimodal distribution, the lower mean pH being 6.74 (+/- 0.23). When 3T3 cells pinocytose medium containing fluorescein dextran, pinosomes peripheral to the nucleus exhibit a lower pH than those closer to the ruffling edge of the cell. The present image processing system is analyzed for linearity of detection, light scattering artifacts, signal to noise ratio, standard curves, and spatial resolution. The results obtained from digital image analysis are shown to be comparable to the results from standard microspectrofluorometry. We also discuss several other applications of this ratio imaging technique in cell biology.     

Compact, high performance surface plasmon resonance imaging system

We report the construction and characterization of a new compact surface plasmon resonance imaging instrument. Surface plasmon resonance imaging is a versatile technique for detection, quantification and visualization of biomolecular binding events which have spatial structure. The imager uses a folded light path, wide-field optics and a tilted detector to implement a high performance optical system in a volume 7 in. x 4 in. x 2 in. A bright diode light source and an image detector with fast frame rate and integrated digital signal processor enable real-time averaging of multiple images for improved signal-to-noise ratio. Operating angle of the imager is adjusted by linear translation of the light source. Imager performance is illustrated using resolution test targets, refractive index test solutions, and competition assays for the antiepileptic drug phenytoin. Microfluidic flowcells are used to enable simultaneous assay of three sample streams. Noise level of refractive index measurements was found to decrease proportional to the square root of the number of pixels averaged, reaching approximately 5 x 10(-7) refractive index units root-mean-square for 160 x 120 pixels image regions imaged for 1s. The simple, compact construction and high performance of the imager will allow the device to be readily applied to a wide range of applications.     

Automated enumeration of groups of marine picoplankton after fluorescence in situ hybridization

We describe here an automated system for the counting of multiple samples of double-stained microbial cells on sections of membrane filters. The application integrates an epifluorescence microscope equipped with motorized z-axis drive, shutters, and filter wheels with a scanning stage, a digital camera, and image analysis software. The relative abundances of specific microbial taxa are quantified in samples of marine picoplankton, as detected by fluorescence in situ hybridization (FISH) and catalyzed reporter deposition. Pairs of microscopic images are automatically acquired from numerous positions at two wavelengths, and microbial cells with both general DNA and FISH staining are counted after object edge detection and signal-to-background ratio thresholding. Microscopic fields that are inappropriate for cell counting are automatically excluded prior to measurements. Two nested walk paths guide the device across a series of triangular preparations until a user-defined number of total cells has been analyzed per sample. A backup autofocusing routine at incident light allows automated refocusing between individual samples and can reestablish the focal plane after fatal focusing errors at epifluorescence illumination. The system was calibrated to produce relative abundances of FISH-stained cells in North Sea samples that were comparable to results obtained by manual evaluation. Up to 28 preparations could be analyzed within 4 h without operator interference. The device was subsequently applied for the counting of different microbial populations in incubation series of North Sea waters. Automated digital microscopy greatly facilitates the processing of numerous FISH-stained samples and might thus open new perspectives for bacterioplankton population ecology.     

Mechanical quality assurance using light field for linear accelerators with camera calibration

Mechanical Quality Assurance (QA) is important to assure spatially precise delivery of external-beam radiation therapy. As an alternative to the conventional-film based method, we have developed a new tool for mechanical QA of LINACs which uses a light field rather than radiation. When light passes through the collimator, a shadow is projected onto a piece of translucent paper and the resulting image is captured by a digital camera via a mirror. With this method, we evaluated the position of the LINAC isocenter and the accuracy of the gantry, collimator, and couch rotation. We also evaluated the accuracy of the digital readouts of the gantry, collimator, and couch rotation. In addition, the treatment couch position indicator was tested. We performed camera calibration as an essential pre-requisite for quantitative measurements of the position of isocenter, the linear motion of the couch, and the rotation angles of the gantry and collimator. Camera calibration reduced the measurement error to submillimeter based on uncertainty in pixel size of the image, while, without calibration, the measurement error of up to 2 mm could occur for an object with a length of 5 cm.     

A microcomputer system for video image analysis and diagnostic microdensitometry

A system for microdensitometry based on a microcomputer, video digitizer and solid-state camera has been developed. Image analysis and densitometry are achieved with convenient control over image editing and calibration. The linear photometric properties of the imaging device enable measurements of high accuracy. The system has proven to give rapid and repeatable performance for determining DNA content distribution from measurements of Feulgen-stained cell nuclei. The results show that a practical image analysis microdensitometer can be designed using a readily available microcomputer. The low cost and simple operation are of benefit for diagnostic applications in which flow cytometry is not possible, the time required for microscope photometry is too great or an automated image analyzer and support staff are not available.     

利用数码相机和Photoshop软件非破坏性测定叶面积的简便方法

采用数码相机获取叶片的数字图像,用Photoshop图像处理软件计算叶面积,并与目前常用的剪纸法和叶面积仪测定法进行比较分析。结果表明,本方法和上述传统测定方法测定结果存在极显著线性相关;不同拍摄分辨率、单位叶面积存贮像素个数和拍摄角度对测定结果无显著影响。和其它方法相比,本方法具有准确、快速、成本低廉、适合非破坏性动态连续观测等优点,适用于植物生理生态学研究中叶面积的测定。     

Measurement of fluorescence using digital integration of video images

The authors describe a system for the quantitation of fluorescence light output by individual cells using the signal obtained from a silicon intensifier target video camera. The video image is digitized to 4 bits (16 levels), and a 512 X 512 matrix is constructed and stored in 128K of video memory. Areas to be measured are user-specified by means of a light pen entry system. Recorded intensities are integrated under microprocessor control to provide a measure of total fluorescence in the selected areas. The distribution of light intensities per pixel over the 16-level gray scale as well as morphometric data are also obtainable. Linear response to transmission and fluorescence standards was verified. Reproducibility of the system was evaluated using fluorescent beads and glutaraldehyde-fixed chick red blood cells, which gave coefficients of variation comparable to those obtainable from other systems. Measurements of DNA per nucleus of human diploid fibroblasts using Hoechst dye 33258 yielded the two sharp peaks corresponding to the 2C and 4C values of DNA expected from such cells. We conclude that digital video measurement of fluorescence provides meaningful data and has considerable promise as a sensitive tool for the quantitation of fluorescence at the cellular level.     

Feasibility of an in situ measurement device for bubble size and distribution

The feasibility of in situ measurement device for bubble size and distribution was explored. A novel in situ probe measurement system, the EnviroCam™, was developed. Where possible, this probe incorporated strengths, and minimized weaknesses of historical and currently available real-time measurement methods for bubbles. The system was based on a digital, high-speed, high resolution, modular camera system, attached to a stainless steel shroud, compatible with standard Ingold ports on fermenters. Still frames and/or video were produced, capturing bubbles passing through the notch of the shroud. An LED light source was integral with the shroud. Bubbles were analyzed using customized commercially available image analysis software and standard statistical methods. Using this system, bubble sizes were measured as a function of various operating parameters (e.g., agitation rate, aeration rate) and as a function of media properties (e.g., viscosity, antifoam, cottonseed flour, and microbial/animal cell broths) to demonstrate system performance and its limitations. For selected conditions, mean bubble size changes qualitatively compared favorably with published relationships. Current instrument measurement capabilities were limited primarily to clear solutions that did not contain large numbers of overlapping bubbles.     

Portable telepathology: methods and tools

Telepathology is becoming easier to implement in most pathology departments. In fact e-mail image transmit can be done from almost any pathologist as a simplistic telepathology system. We tried to develop a way to improve capabilities of communication among pathologists with the idea that the system should be affordable for everybody. We took the premise that any pathology department would have microscopes and computers with Internet connection, and selected a few elements to convert them into a telepathology station. Needs were reduced to a camera to collect images, a universal microscope adapter for the camera, a device to connect the camera to the computer, and a software for the remote image transmit. We found out a microscope adapter (MaxView Plus) that allowed us connect almost any domestic digital camera to any microscope. The video out signal from the camera was sent to the computer through an Aver Media USB connector. At last, we selected a group of portable applications that were assembled into a USB memory device. Portable applications are computer programs that can be carried generally on USB flash drives, but also in any other portable device, and used on any (Windows) computer without installation. Besides, when unplugging the device, none of personal data is left behind. We selected open-source applications, and based the pathology image transmission to VLC Media Player due to its functionality as streaming server, portability and ease of use and configuration. Audio transmission was usually done through normal phone lines. We also employed alternative videoconferencing software, SightSpeed for bi-directional image transmission from microscopes, and conventional cameras allowing visual communication and also image transmit from gross pathology specimens. All these elements allowed us to install and use a telepathology system in a few minutes, fully prepared for real time image broadcast.     

A polychromator-based microspectrophotometer

A microspectrophotometer is a digital microscope used to measure absorption and fluorescence spectra. In this paper we describe a polychromator-based microspectrophotometer that performs in vivo absorption or emission measurements at the same time on different subcellular compartments such as photoreceptive and photosynthetic structures of algal cells. In this system, a flat field imaging concave grating polychromator is connected to the slit-shaped exit pupil of a light-guide probe mounted onto a microscope equipped with an epifluorescence module. The subcellular components, on which the spectra will be measured, are placed in the microscope field and finely adjusted. The outer bundle of the probe is used for centering the objects, while the central bundle of the probe, containing 19 light guides, is used for acquiring either transmitted or emitted light (i.e. fluorescence). The light transmitted or emitted by the subcellular components is collected by the probe mounted in the back focal plane of the ocular. The exit pupil of this probe, connected to a flat field imaging concave grating polychromator, produces a dispersion image that in turn is focused onto a digital slow scan cooled CCD camera. Absorption and emission spectra of algal subcellular compartments are presented.     

Performance evaluation of a machine vision system for insect pests identification of field crops using artificial neural networks

Feature extraction is a crucial part of advanced image recognition systems. In this research, an autonomous detection device was designed and developed for insect pest detection to improve the ability of intelligent systems in order to annihilate harmful insect pests in agricultural crop fields. Device included a dark chamber, a CCD digital camera, a LDR lightening module and a personal computer. The proposed programme for precise insect pest detection was based on an image processing algorithm and artificial neural networks (ANNs). After image acquisition, the insect pests’ images were extracted from original images with Canny filtration. Afterwards, four morphological and three textural features from the obtained images were measured and normalised. Performance of ANN model was tested successfully for Beet armyworm (Spodoptera exigua) recognition in images using back-propagation supervised learning method and inspection data. Results showed that proposed system was able to identify S. exigua in the images from other species. Such this machine vision system can be used in autonomous field robots to achieve a modern farmer’s assistant.     

Quantitative Assessment of Lesion Characteristics and Disease Severity Using Digital Image Processing

This work describes a dedicated software which detects and characterizes disease lesions on leaves to provide data on the number and type of lesions and the percentage of leaf area diseased (severity). The software, written in C²⁺, can be used with a standard computer in combination with a colour CCD camera and a frame grabber for image acquisition. The usefulness and adaptability of the software was evaluated using two foliar diseases, Alternaria blight of sunflower and oat leaf rust (Puccinia coronata f.sp. avenae), which differ in symptoms. Using image segmentation and classification techniques, the software discriminated disease symptoms from the healthy leaf area. The number and size of lesions and severity, obtained using the image processing software, were compared with those calculated using a software planimeter or visual assessment. Significant linear relationships between planimeter and the imaging software were obtained for lesion number and severity in oat leaf rust and for severity in sunflower blight. Artefacts, mistakenly classified as blight lesions by the imaging software resulted in an over-estimation of the number of lesions. Future research is aimed at improving accuracy through better illumination during image capture. A dedicated, compact and portable hardware is currently being developed for field use as a self-contained device for disease assessment.