Machine Vision Analysis Research Articles

Chemical address tags of fluorescent bioimaging probes

Chemical address tags can be defined as specific structural features shared by a set of bioimaging probes having a predictable influence on cell-associated visual signals obtained from these probes. Here, using a large image dataset acquired with a high content screening instrument, machine vision and cheminformatics analysis have been applied to reveal chemical address tags. With a combinatorial library of fluorescent molecules, fluorescence signal intensity, spectral, and spatial features characterizing each one of the probes' visual signals were extracted from images acquired with the three different excitation and emission channels of the imaging instrument. With multivariate regression, the additive contribution from each one of the different building blocks of the bioimaging probes toward each measured, cell-associated image-based feature was calculated. In this manner, variations in the chemical features of the molecules were associated with the resulting staining patterns, facilitating quantitative, objective analysis of chemical address tags. Hierarchical clustering and paired image-cheminformatics analysis revealed key structure-property relationships amongst many building blocks of the fluorescent molecules. The results point to different chemical modifications of the bioimaging probes that can exert similar (or different) effects on the probes' visual signals. Inspection of the clustered structures suggests intramolecular charge migration or partial charge distribution as potential mechanistic determinants of chemical address tag behavior.

Review of Stereo Vision Algorithms: From Software to Hardware

Stereo vision, resulting in the knowledge of deep information in a scene, is of great importance in the field of machine vision, robotics and image analysis. In this article, an explicit analysis of the existing stereo matching methods, up to date, is presented. The presented algorithms are discussed in terms of speed, accuracy, coverage, time consumption, and disparity range. Towards the direction of real-time operation, the development of stereo matching algorithms, suitable for efficient hardware implementation is highly desirable. Implementations of stereo matching algorithms in hardware for real-time applications are also discussed in details.

Comparison of an Automated Pattern Analysis Machine Vision Time-lapse System with Traditional Endpoint Measurements in the Analysis of Cell Growth and Cytotoxicity

Machine vision is an application of computer vision. It both collects visual information and interprets the images. Although the machine obviously does not 'see' in the same sense that humans do, it is possible to acquire visual information and to create programmes to identify relevant image features in an effective and consistent manner. Machine vision is widely applied in industrial automation, but here we describe how we have used it to monitor and interpret data from cell cultures. The machine vision system used (Cell-IQ) consisted of an inbuilt atmosphere-controlled incubator, where cell culture plates were placed during the test. Artificial intelligence (AI) software, which uses machine vision technology, took care of the follow-up analysis of cellular morphological changes. Basic endpoint and staining methods to evaluate the condition of the cells, were conducted in parallel to the machine vision analysis. The results showed that the automated system for pattern analysis of morphological changes yielded comparable results to those obtained by conventional methods. The inbuilt software analysis offers a promising way of evaluating cell growth and various cell phases. The continuous follow-up and label-free analysis, as well as the possibility of measuring multiple parameters simultaneously from the same cell population, were major advantages of this system, as compared to conventional endpoint measurement methodology.

Advanced ATR techniques based on High-Resolution SAS Sensors

Automatic Target Recognition (ATR) is a key element of expeditionary Mine Countermeasures (MCM) and port protection operations. Most existing approaches to ATR are currently based on high resolution sonar sensors, which provide enough information to obtain satisfactory detection and classification performance for the large World War 2 mine types (e.g., 2 m long cylinder). False alarm rates, however, are still unacceptably high for modern mines, which constrains the way operations are undertaken and often requires either confirmation or re-evaluation by a human operator. The introduction of new AUV-mounted Synthetic Aperture Sonars (SAS) increases the resolution, quality and range of acquired sonar images, which broadens the set of machine vision and computer image analysis techniques that can be used for underwater ATR operations. In this paper we study the impact that the increased quality and resolution have on performance gains and false alarm reduction. A number of classification algorithms are selected to represent the pool of existing approaches to target detection and classification, and their performances are estimated using both simulated and real image data in order to quantify the benefits associated to the new SAS technology. Evolution and near-future plans are discussed, introducing emerging bio-sonar sensors, anomaly detectors and autonomous AUV systems.

Machine vision analysis of the energy efficiency of intermodal freight trains

Intermodal (IM) trains are typically the fastest freight trains operated in North America. The aerodynamic characteristics of many of these trains are often relatively poor resulting in high fuel consumption. However, considerable variation in fuel efficiency is possible depending on how the loads are placed on railcars in the train. Consequently, substantial potential fuel savings are possible if more attention is paid to the loading configuration of trains. A wayside machine vision (MV) system was developed to automatically scan passing IM trains and assess their aerodynamic efficiency. MV algorithms are used to analyse these images, detect and measure gaps between loads. In order to make use of the data, a scoring system was developed based on two attributes - the aerodynamic coefficient and slot efficiency. The aerodynamic coefficient is calculated using the Aerodynamic Subroutine of the train energy model. Slot efficiency represents the difference between the actual and ideal loading configuration given the particular set of railcars in the train. This system can provide IM terminal managers feedback on loading performance for trains and be integrated into the software support systems used for loading assignment.

An intelligent emergency response system: Preliminary development and testing of a functional health monitoring system

Changes in a person’s routine of daily activities can signal a change in health. To support the growing elderly popu-lation who want to age-in-place, techniques and algorithms have been developed to build a system that monitors functional health in the home environment. This health monitoring system has been developed with machine vision and pattern analysis components to track the occupant, learn his/her pattern of activity, and detect significant deviations that could indicate a change in health status. The ef-fectiveness of the health monitoring system was investigated with a pilot study capturing video footage of a 28-day simulation including 21 days of normal activ-ity and seven days of abnormal scenarios. The system was effective in learning an occupant’s pattern of activity and detecting deviations that were indicative of changes in the occupant’s functional health status. Overall, the results indicate that a health monitoring system could be developed that uses machine vision and basic artificial intelligence with promising potential to support aging-in-place.

Evaluation of Color Parameters in a Machine Vision Analysis of Carbon Monoxide-Treated Fish—Part I

ABSTRACT The color of fresh tuna treated by 4% carbon monoxide +20% carbon dioxide +10% oxygen, or irradiated at 1 KGy or 2 KGy, or first gas treated then irradiated, was analyzed using the R, a* and hue parameters in a machine vision system. The percent of sample surface having R > Rthreshold, a* < a*threshold, or hue < huethreshold was associated with storage time for different levels of threshold values. Hue was selected as the best representative of the red color of tuna. A method was suggested to select the threshold value of a parameter. Irradiation did not change the color of fresh tuna. Exposure to CO increased the redness, and preserved it for up to 12 days in refrigerated storage. Microbial numbers for CO treated tuna were not different from control. Irradiation at 2 KGy was very effective in reducing microorganisms, while 1 KGy was ineffective.

Machine Vision Analysis of Antibrowning Potency for Oxalic Acid: A Comparative Investigation on Banana and Apple

ABSTRACT: Relative antibrowning potency of oxalic acid on banana and apple slices was investigated using a machine vision system. Degree of browning on fresh‐cut surfaces was evaluated visually and quantitatively by observing changes of CIE L* values and evaluating temporal changes in color spectra based on experimental variables, oxalic acid concentration, and storage time. Browning inhibition was most prominent on banana and apple slices treated with oxalic acid solutions at concentrations of 60 and 5 mM, respectively. Oxalic acid was a more potent antibrowning agent compared with other structurally related acids. Average residual oxalic acid levels in the tissues for an effective antibrowning activity were measured.

AE—Automation and Emerging Technologies: Automatic Mushroom Harvester Development

This paper describes research which demonstrates that robotic harvesting of high-quality mushrooms, for the fresh market, is entirely feasible. The work was undertaken in two stages. First, systems capable of automatic mushroom location, sizing, selection, picking, trimming, conveying and transfer were conceived and implemented on a specially designed laboratory development rig. Once the sub-systems and integrated control software had been proven, the second stage of producing a pilot robotic harvester, based on these ideas, was carried out. Machine vision and image analysis algorithms were used for mushroom location and sizing. A separate strategy algorithm was written for selecting the most appropriate mushroom picking order. Overall system control was achieved using a single 486 DX33 MHz personal computer. Separate suction cup end-effectors were produced for picking in confined (less than 300 mm) or open headroom spaces. New delicate handling techniques for conveying, trimming and transferring mushrooms using flexible fingers, high-speed knives and a padded pneumatic gripper system, were developed. Both the laboratory rig and the pilot harvester were evaluated using imitation mushrooms constructed from dental plaster and a wide range of cultivated Agaricus bisporus mushrooms. In trials on a commercial farm, an average of over 80% of the mushroom pick attempts were successful.

SURFACE MATCHING AND DIFFERENCE DETECTION WITHOUT THE AID OF CONTROL POINTS

If photogrammetry is to be applied to applications such as machine vision, industrial inspection, or medical analysis, more diverse three-dimensional model matching procedures need to be developed. Further, automatic surface difference detection should also be incorporated into the derived algorithms in order to accommodate the requirements of these new applications.This paper reviews a wide range of literature and presents the current status of three-dimensional model matching. A comparison is made between two-dimensional and three-dimensional matching for the purpose of delineating and similarities and differences. Finally, an algorithm developed by the author for surface matching is also discussed.

SURFACE MATCHING AND DIFFERENCE DETECTION WITHOUT THE AID OF CONTROL POINTS

Abstract If photogrammetry is to be applied to applications such as machine vision, industrial inspection, or medical analysis, more diverse three-dimensional model matching procedures need to be developed. Further, automatic surface difference detection should also be incorporated into the derived algorithms in order to accommodate the requirements of these new applications.This paper reviews a wide range of literature and presents the current status of three-dimensional model matching. A comparison is made between two-dimensional and three-dimensional matching for the purpose of delineating and similarities and differences. Finally, an algorithm developed by the author for surface matching is also discussed.

Determining Detailed Origin-Destination and Travel Time Patterns Using Video and Machine Vision License Plate Matching

Examples are presented of the use of machine vision technology for reading and processing vehicle license plate images acquired in the field by video camcorders. Applications include travel time and microscale origin-destination pattern analyses for a freeway interchange in Tampa and the road network serving Houston's Intercontinental Airport, and high-occupancy vehicle lane performance measurements in Seattle. It is concluded that video and machine vision analysis of license plates is an effective and efficient means of conducting a wide variety of traffic engineering and traffic management studies.

Determining Detailed Origin-Destination and Travel Time Patterns Using Video and Machine Vision License Plate Matching

Examples are presented of the use of machine vision technology for reading and processing vehicle license plate images acquired in the field by video camcorders. Applications include travel time and microscale origin-destination pattern analyses for a freeway interchange in Tampa and the road network serving Houston's Intercontinental Airport, and high-occupancy vehicle lane performance measurements in Seattle. It is concluded that video and machine vision analysis of license plates is an effective and efficient means of conducting a wide variety of traffic engineering and traffic management studies.

Integrating Photogrammetric Techniques With Machine Vision and Scene Analysis II

Integrating Photogrammetric Techniques With Machine Vision And Scene Analysis II by I. J. Dowman, p.323Kodak Digital Cameras For Small Format Aerial Photography by R. W. Graham, p.325From Pixels To Sequences by S. Robson, p.327

Non-destructive machine vision analysis of pigment-producing cell cultures

Macroscopic images of callus colonies and microscopic images from suspension cultures of anthocyanin-producing Ajuga cells were analyzed by machine vision to appraise culture productivity (pigment accumulation and cell growth). Callus culture mass was accurately estimated using data from simultaneous capture of top and side view images. A comparison of several alternative regression models determined that this performance data could be obtained reliably without interrupting the culture cycle to weigh cell colonies. The best regression models could predict callus mass in the 2.12–5.95 g fresh weight range with a maximum error less than 11%. For suspension culture analysis, the detailed RGB (red-green-blue) information captured in each of the color microscopic images did not allow separation of pigmented cells and cell aggregates from non-pigmented entities and background in the medium, however, conversion of RGB data to the HSI (hue-saturation-intensity) coordinate system permitted clear separation of object classes based on a combination of dimensional and photometric information. Further segmentation using saturation and intensity characteristics of the HSI values allowed categorization of low, medium, and highly-pig- mented cells and aggregates in the mixed suspension, and machine vision data was able to track both biomass accumulation and anthocyanin pigment formation over time as verified by mass and spectrophotometric analysis of the same cultures.

Automatic parsing and indexing of news video

Automatic construction of content-based indices for video source material requires general semantic interpretation of both images and their accompanying sounds; but such a broadly-based semantic analysis is beyond the capabilities of the current technologies of machine vision and audio signal analysis. However, if one can assume a limited and well-demarcated body of domain knowledge for describing the content of a body of video, then it becomes easier to interpret a video source in terms of that domain knowledge. This paper presents our work on using domain knowledge to parse news video programs and to index them on the basis of their visual content. Models based on both the spatial structure of image frames and the temporal structure of the entire program have been developed for news videos, along with algorithms that apply these models by locating and identifying instances of their elements. Experimental results are also discussed in detail to evaluate both the models and the algorithms that use them. Finally, proposals for future work are summarized.

Machine Vision Analysis of Welding Region and its Application to Seam Tracking in Arc Welding

Unlike the application of machine vision in many other fields, there is a particular problem in developing machine vision for automated and robotic welding processes, because the disturbance of the arc light deteriorates the field to be viewed. This paper describes an analysis of the radiated energy from the weld pool, and based on this describes how vision sensing of the welding region may be improved. An approach using specified pattern parameters is described to evaluate the ability to recognize an acquired image under different conditions such as with vidicons or sensor devices, different welding variables and optical parameters. Based on analysis of radiation from the welding region, a satisfactory wavelength range for sensing the image of the weld region is proposed and proved by experimentation. Images of gas tungsten arc welding (GTAW) and gas metal arc welding (GMAW) regions have been modelled and successfully implemented for seam tracking control with a simultaneously developed low-cost vision system.

LIGHT INFLUENCE ON IN VITRO ANTHOCYANIN PRODUCTION IN THREE CRANBERRY (Vaccinum macrocarpon) CULTIVARS

The influence of spectral irradiance on in vitro anthocyanin production in white friable callus of cranberry cultivars `Stevens', `Searles', and `Ben Lear' was examined by modifying cool white fluorescent lamp emission. Filters were used to shift spectral balance to blue, yellow, and red wavelengths at different PPF levels within a span of 30 to 150 μmol m-2s-1. White, friable, dark-grown calli—originally derived from shoots and leaves of shoot culture stock plants—were maintained approximately 6 months with subculture every 4 wk. Uniform callus masses were then subcultured to pigment induction medium and transferred to treatment microenvironments. Visual and machine vision analysis of pigmentation was assessed weekly for one month. Higher PPF levels stimulated the most rapid anthocyanin synthesis. Calli in the blue treatment produced the greatest pigmentation within the first two weeks. All three cultivars responded similarly to the treatments, however, pigmentation was slightly more intense in `Ben Lear' at higher PPFs. Results indicate spectral irradiance significantly affects anthocyanin synthesis in cranberry callus cultures, and suggest that manipulation of the physical microenvironment is an important consideration in development of efficient in vitro natural pigment production systems.

Potential Methods for Detecting Pits in Tart Cherries

Approximately 98% of the U.S. tart cherry crop is pitted and processed into canned, frozen, and dried fruit or made into juice. U.S. Grade A pitted tart cherries are currently allowed to contain up to 1 pit per 1.1 kg (2.5 lb) of finished cherries. To minimize product liability, and maximize end-user and consumer acceptance of pitted cherries, a finished pack having not over 1 pit per 28.3 kg (62.4 lb) of cherries is desired. Potential methods for the automatic detection and rejection of unpitted cherries and loose pits were investigated. Light transmission coupled with machine vision and image analysis provided evidence for the highest possibility of success at this time.

Intrinsic characteristics as the interface between CAD and machine vision systems

Computer Aided Design systems are currently being used to drive machine vision analysis. We propose that the set of intrinsic 3-D shape characteristics provided by the CAD system be used to compare representational power of different CAD systems so far as the requirements for machine vision are concerned.