Browsing by Subject "Image analysis"
Now showing 1 - 5 of 5
- Results Per Page
- Sort Options
Item 2015 Twin Cities Metropolitan Area Urban Tree Canopy Assessment(2017-01-03) Knight, Joe F; Rampi, Lian P; Host, Trevor K; jknight@umn.edu; Knight, Joseph, FA high-resolution (1-meter) tree canopy assessment was completed for the Twin Cities Metropolitan Area. Mapping of existing and potential tree canopy is critical for urban tree management at the landscape level. This classification was created from combined 2015 aerial imagery, LIDAR data, and ancillary thematic layers. These data sets were integrated using an Object-Based Image Analysis (OBIA) approach through multi-resolution image segmentation and an iterative set of classification commands in the form of customized rulesets. eCognition® Developer was used to develop the rulesets and produce raster classification products for TCMA. The results were evaluated using randomly placed and independent verified assessment points. The classification product was analyzed at regional scales to compare distributions of tree canopy spatially and at different resolutions. The combination of spectral data and LiDAR through an OBIA method helped to improve the overall accuracy results providing more aesthetically pleasing maps of tree canopy with highly accurate results.Item Deployment of Practical Methods for Counting Bicycle and Pedestrian Use of a Transportation Facility(Intelligent Transportation Systems Institute, Center for Transportation Studies, 2012-01) Somasundaram, Guruprasad; Morellas, Vassilios; Papanikolopoulos, NikolaosThe classification problem of distinguishing bicycles from pedestrians for traffic counting applications is the objective of this research project. The scenes that are typically involved are bicycle trails, bridges, and bicycle lanes. These locations have heavy traffic of mainly pedestrians and bicyclists. A vision-based system overcomes many of the shortcomings of existing technologies such as loop counters, buried pressure pads, infra-red counters, etc. These methods do not have distinctive profiles for bicycles and pedestrians. Also most of these technologies require expert installation and maintenance. Cameras are inexpensive and abundant and are relatively easy to use, but they tend to be useful as a counting system only when accompanied by powerful algorithms that analyze the images. We employ state-of-the-art algorithms for performing object classification to solve the problem of distinguishing bicyclists from pedestrians. We detail the challenges that are involved in this particular problem, and we propose solutions to address these challenges. We explore common approaches of global image analysis aided by motion information and compare the results with local image analysis in which we attempt to distinguish the individual parts of the composite object. We compare the classification accuracies of both approaches on real data and present detailed discussion on practical deployment factors.Item Physical and Image Analysis Sizing of Mine Run Taconite Ore(University of Minnesota Duluth, 1993-10) Niles, Harlan BThe United States Bureau of Mines (USBM) has developed an image analysis system to determine the size distribution of mine-run taconite ore. The results can be used to evaluate fragmentation so that blasts can be designed for better productivity. ~he Coleraine Minerals Research Laboratory (CMRL) has been testing the system at Minntac with apparent success, but there was no method available for testing the accuracy of size distributions determined by image analysis. The Iron Ore Cooperative Research Committee approved and funded a project to screen mine-run ore and compare the results to sizing by image analysis. Three samples, about 2500 tons each, were sized on 6 and 12 inches in a contractor's vibrating grizzly plant at Minntac. Size analyses were extended to 65 mesh by sizing samples of minus-6 inch at the Coleraine laboratory. Between 30 and 60 plus- 12 inch pieces from each bulk sample were measured to provide thickness-width-length aspect ratios and to indicate large-fragment dimensions and weights that are encountered in mine-run ore. The average ratio of thickness to width to length was 1.00:2.20:3.21. Fragment volumes ranged from 4 to 80 cubic feet. Weights of these would be approximately 800 pounds and 16,000 pounds. After the size fractions had been weighed, they were recombined, each sample was loaded into rail cars, and they were dumped at number one primary crusher. The USBM and CMRL video-taped the ore as it was dumped and processed the tapes through a computer to produce an image analysis size distribution for each of the three samples. Because of a large discrepancy between the contractor's and the track scale weights, the size analysis of sample number one was not acceptable. The physical size analyses of samples two and three and their corresponding image analysis size distributions were nearly identical for fragment widths of at least 12 inches. Size. distributions of mine-run ore by the image analysis system have been proven reliable for evaluating the effective fragmentation of individual blasts for sizes down to 12 inches in width.Item Pragmatic Methods for Perennial Ryegrass (Lolium perenne L.) Breeding(2019-09) Heineck, GarettThis dissertation consists of five chapters, each was written as a stand-alone manuscript. The introduction of each chapter serves, in part, as the introductory literature review. Herein is described a brief summary of each chapter’s introduction, methods, and results. Chapter 1 Crown and stem rust are major diseases of perennial ryegrass (Lolium perenne L.). Plant breeders and pathologists often rate rust severity in the field using the modified Cobb scale, but this method is subjective and labor intensive. A novel, open-source system using ImageJ and R was developed to quantify pustule number and area using digital images collected from spaced plants in the field. The computer-processing pipeline included development of training data for prediction of pixel identity using random forest and noise reduction spatial processing. Raters and the computer scored rust severity on plant images of varying complexity including whole-plant (WP), five-leaf (FL), and single-leaf (SL) image series. Computer accuracy was determined using the SL, while the FL series gave insight into the true value of WP severity. Rater ability was assessed using a panel of nine scientists with varying levels of disease rating experience. Results showed rater perceptions of crown rust severity were very consistent across images, but agreement on severity values for a given image were low. Rater consistency for stem rust severity was low and FL scores were not strongly correlated with WP scores (r=0.36, P=0.03) indicating low rater accuracy. The computer-processing pipeline was able to accurately discriminate, count and quantify crown and stem rust pustules on leaf samples. Correlations between computer and the median rater score for crown rust were excellent (r>0.90, P< 0.001) for all image series. Similar to raters, there was lack of correlation between WP and FL series (r=0.20, NS) indicating this technique is limited to leaf or stem samples for stem rust and not applicable to WP. However, the computer-processing pipeline shows promise in replacing visual rating of WP for crown rust. Chapter 2 Perennial ryegrass is an important turf and forage species that often becomes infected with crown rust caused by Puccinia coronata f. sp. lolii. Disease control through Clavicipitaceous endophytes has been proposed as a potential biocontrol. Two field experiments were designed to determine the influence of native Epichloë endophyte infection on natural rust infection across a diverse panel of perennial ryegrass germplasm. Experiment 1 used an isogenic population design in which clonal plants infected (E+) or endophyte free (E-) were nested within 14 perennial ryegrass entries. Experiment 2 consisted of E+ and E- progeny from isogenic parents. Results showed the endophyte had no consistent marginal effect on crown rust severity across or within entries; however, several isogenic host pairs did show either favorable or antagonistic effects. Despite these sporadic effects, no differences were found between isofrequent family pairs indicating the presence of a host by endophyte interaction. These findings support the conclusion that endophyte infection does not play a substantial role in mediating crown rust severity on a population scale. Genotypic and phenotypic data revealed that endophyte isolates were similar within entry indicating that host genotype could be responsible for the highly specific endophyte effect on crown rust. The importance of host genotype was further supported by substantial heritability estimates for disease severity. Chapter 3 Understanding trade-offs between breeding for turfgrass performance and seed production capacity is beneficial for turfgrass breeders working on the improvement of perennial ryegrass. An experiment was designed to identify potential tradeoffs and their mechanistic causes by measuring turfgrass and seed production traits on 20 perennial ryegrass entries grown in two Minnesota environments. Average turfgrass quality scores were not correlated with seed yield at either location. However, several individual turf quality rating dates were moderately correlated with seed yield at both locations (P < 0.1). Within these dates, subsidiary turfgrass traits were exhaustively regressed against quality data to identify the optimum combination of variables explaining turfgrass quality. Turf quality was driven by lateral and vertical growth rate, crown rust severity, winter survival, and stemminess. Of these traits it was clear that breeding for increased lateral growth rate and winter survival would be positively correlated with increased yield. Crown rust severity was nearly perfectly correlated across environments (r > 0.95) indicating a favorable response. Vertical growth rate and fertile tiller production; however, were negatively associated with turf quality and were investigated to discover whether breeding for slower growth and steminess would negatively influence seed yield. Although fertile tillers were associated with higher seed yield, there was no relationship with steminess in turfgrass plots. Vertical growth rate was associated with earlier maturity, which is associated with increased seed yield indicating a slight tradeoff. Evidence from the environments and germplasm tested suggest that plant breeders should generally not be concerned with negative tradeoffs between perennial ryegrass turf quality and seed yield. Chapter 4 Producing adequate seed yield is essential for perennial ryegrass cultivar success in both the turf and forage industries. This study examined the importance of seed yield components across 20 perennial ryegrass entries in both spaced plantings and seed production swards at two locations in Minnesota. Competitive (23 plants m-2) and non-competitive (3 plants m-2) spaced plant nurseries were tested for predictive ability. Structural equation modeling (SEM) was used to determine the indirect and direct influences of yield components on total seed yield. The impetus of this approach was to discover new breeding targets and an ideotype for increasing seed yield. Results showed that when tiller survival was low winterkill was highly influential on seed yield both directly and indirectly through fertile tiller number. Fertile tiller number was more important in spaced plant environments than in swords, but very few differences were found between entries. Spike fertility directly influenced spike yield and was indirectly important for total seed yield. Although the relative importance of individual seed yield components was similar between nursery designs, the competitive design had a superior predictive ability for sward yield via total plant yield and spike fertility. Chapter 5 Turf-type perennial ryegrass success depends both on adequate turfgrass quality, but also economic seed yield. In most breeding programs, spaced plants are the initial unit of selection in which observations of related individuals dictate selections of superior germplasm for further testing. As such, spaced plants must be predictive of both seed production and turfgrass growing environments. This research investigated the effectiveness of both standard (3 plants m-2) and competitive (23 plants m-2) spaced plant nurseries as selection environments with respect to sward environments using the same 20 turf-type entries. Seed production, turfgrass and the two spaced plant growing environments were tested at two locations in Minnesota. Turfgrass quality traits were measured in 2017 and 2018 and seed production traits were measured in 2018. Rank correlations for fertility index between the competitive nursery and sward environments at both locations was substantial (rp = 0.52 and 0.81, P < 0.05). Genetic color and crown rust severity were the most prominent variables in models for predicting turfgrass quality. Competition between spaced plants made only a minor improvement of predictive ability for these traits. Overall, increased competition between spaced plants increased the predicative ability (rs) for both turfgrass and seed production traits. Furthermore, the competitive design takes up less space and often makes measurements and observations much easier for bunch-type grasses.Item Uncovering Hidden Phenotypes to Accelerate Domestication in Perennial Ryegrass for Seed Production(2022-10) Barreto Ortiz, JoanSeed dispersion, shattering, shedding, and lack of retention, all refer to the same dispersal mechanism in which reproductive organs detach from plant inflorescences upon maturity. This separation, or disarticulation, is mainly attributed to the development of an abscission layer which is genetically programmed. However, such detachment is just part of the seed dispersal phenomenon determined by a network of interactions between biotic and abiotic agents allowing gene flow over space and time. Understanding the genetics underlying seed dispersal has ecological and agricultural implications; while this phenomenon grants an evolutionary advantage to wild plants, it affects crop productivity and domestication of species with economic potential by severely reducing seed yield. In consequence, variation for dispersal-related traits has been selected over millennia through improving seed retention and yield, thus allowing the development of agriculture and consequent human societies. Nevertheless, genetic correlations among traits can limit the progress of selecting for a dispersal trait and generate unfavorable trade-offs. The presence of these correlations can be attributed to (i) the univariate approach that characterizes both selection methods and the study of multivariate phenotypes like inflorescence architecture and seed dispersal, and (ii) our limited ability to perceive and quantify the multidimensional phenotypic reality. These are the two foci of this introductory chapter, in which I use perennial ryegrass (Lolium perenne L.) as an example to discuss the need for a holistic understanding of the relationships between inflorescence morphology and seed dispersal, with an ultimate goal to improve seed yield and plant domestication.