THE DESIGN AND ANALYSIS OF EXPERIMENTS. By Oscar Kempthorne. New York, John Wiley and Sons, Inc., 1952. 631 pp. $8.50. This book by a teacher of statistics (as well as a consultant for \"experimenters\") is a comprehensive study of the philosophical background for the statistical design of experiment. It is necessary to have some facility with algebraic notation and manipulation to be able to use the volume intelligently. The problems are presented from the theoretical point of view, without such practical examples as would be helpful for those not acquainted with mathematics. The mathematical justification for the techniques is given. As a somewhat advanced treatment of the design and analysis of experiments, this volume will be interesting and helpful for many who approach statistics theoretically as well as practically. With emphasis on the \"why,\" and with description given broadly, the author relates the subject matter to the general theory of statistics and to the general problem of experimental inference. MARGARET J. ROBERTSON

The Design and Analysis of Experiments

物件導向軟體之架構(Object-Oriented Software Construction)探討

The intensity of the infrared radiation emitted by objects is mainly a function of their temperature. In infrared thermography, this feature is used for multiple purposes: as a health indicator in medical applications, as a sign of malfunction in mechanical and electrical maintenance or as an indicator of heat loss in buildings. This paper presents a review of infrared thermography especially focused on two applications: temperature measurement and non-destructive testing, two of the main fields where infrared thermography-based sensors are used. A general introduction to infrared thermography and the common procedures for temperature measurement and non-destructive testing are presented. Furthermore, developments in these fields and recent advances are reviewed.

/pdf/infrared-thermography-for-temperature-measurement-and-non-2r2amijcr4.pdf

Infrared thermography for temperature measurement and non-destructive testing.

A noise robust method based on completed local binary patterns for hot-rolled steel strip surface defects

Automated surface-anomaly detection using machine learning has become an interesting and promising area of research, with a very high and direct impact on the application domain of visual inspection. Deep-learning methods have become the most suitable approaches for this task. They allow the inspection system to learn to detect the surface anomaly by simply showing it a number of exemplar images. This paper presents a segmentation-based deep-learning architecture that is designed for the detection and segmentation of surface anomalies and is demonstrated on a specific domain of surface-crack detection. The design of the architecture enables the model to be trained using a small number of samples, which is an important requirement for practical applications. The proposed model is compared with the related deep-learning methods, including the state-of-the-art commercial software, showing that the proposed approach outperforms the related methods on the specific domain of surface-crack detection. The large number of experiments also shed light on the required precision of the annotation, the number of required training samples and on the required computational cost. Experiments are performed on a newly created dataset based on a real-world quality control case and demonstrates that the proposed approach is able to learn on a small number of defected surfaces, using only approximately 25–30 defective training samples, instead of hundreds or thousands, which is usually the case in deep-learning applications. This makes the deep-learning method practical for use in industry where the number of available defective samples is limited. The dataset is also made publicly available to encourage the development and evaluation of new methods for surface-defect detection.

Segmentation-based deep-learning approach for surface-defect detection

Flatness is a major geometrical feature of rolled products specified by both production and quality needs. Real-time inspection of flatness is the basis of automatic flatness control. Industrial facilities where rolled products are manufactured have adverse environments that affect artificial vision systems. We present a low-cost flatness inspection system based on optical triangulation by means of a laser stripe emitter and a CMOS matrix camera, designed to be part of an online flatness control system. An accurate and robust method to extract a laser stripe in adverse conditions over rough surfaces is proposed and designed to be applied in real time. Laser extraction relies on a local and a global search. The global search is based on an adjustment of curve segments based on a split-and-merge technique. A real-time recording method of the input data of the flatness inspection system is proposed. It stores information about manufacturing conditions for an offline tuning of the laser stripe extraction method using real data. Flatness measurements carried out over steel strips are evaluated quantitatively and qualitatively. Moreover, the real-time performance of the proposed system is analyzed.

/pdf/real-time-flatness-inspection-of-rolled-products-based-on-1vmypnf3n2.pdf

Real-time flatness inspection of rolled products based on optical laser triangulation and three-dimensional surface reconstruction

An improved 3D imaging system for dimensional quality inspection of rolled products in the metal industry

Defect detection is becoming an increasingly important task during the manufacturing process. The early detection of faults or defects and the removal of the elements that may produce them are essential to improve product quality and reduce the economic impact caused by discarding defective products. This point is especially important in the case of products that are very expensive to produce. In this paper, the authors propose a method to detect a specific type of defect that may occur during the production of web materials: periodical defects. This type of defect is very harmful, as it can generate many surface defects, greatly reducing the quality of the end product and, on occasions, making it unsuitable for sale. To run the proposed method, two different functions must be executed a large number of times. Since the time available to perform the detection of these defects may be limited, it is very important to consume the least amount of time possible. In order to reduce the overall time required for detection, an analysis of how the method accesses the input data is performed. Thus, the most efficient data structure to store the information is determined. At the end of the paper, several experiments are performed to verify that both the proposed method and the data structure used to store the information are the most suitable to solve the aforementioned problem.

An efficient method for defect detection during the manufacturing of web materials

During the production of steel strips, a large amount of surface defects can be generated, due to harsh environmental conditions. A high number of surface defects can lead to rejection by the customer, which represents significant economic losses to the production plant. Thus, it is very important to detect the presence and type of the defects generated during the production of each steel strip. Using this information, it is possible to determine whether a strip is suitable for sale, and it may also be useful to determine the origin of defects and, if possible, prevent them from being generated in subsequent strips. To perform these tasks, non-invasive inspection techniques are usually used, carried out automatically by artificial vision systems. Although the inspection conducted by humans is more accurate, they become fatigued quickly, or may even be unable to carry out the inspection correctly when the forward speed of the strip is high. In this paper, a new detection technique is proposed, based on the division of an image into a set of overlapping areas. The optimum values for the configuration parameters of the detection technique are automatically determined using a genetic algorithm. After the detection phase, all the defects are classified using a neural network. A very satisfactory success rate has been achieved in both detection and classification phases.

Francisco G. Bulnes

Papers

Infrared thermography for temperature measurement and non-destructive testing.

Real-time flatness inspection of rolled products based on optical laser triangulation and three-dimensional surface reconstruction

An improved 3D imaging system for dimensional quality inspection of rolled products in the metal industry

An efficient method for defect detection during the manufacturing of web materials

A Non-Invasive Technique for Online Defect Detection on Steel Strip Surfaces