Ginned cotton’s quality is one significant factor to evaluate the cotton grade and influence the yarn qualities. Ginned cotton is always mixed with contaminants during picking, storing, drying, transporting, purchasing, and processing. Manual evaluation is time consuming, labor intensive, and unreliable. This paper proposed a fast feature extraction algorithm is presented for the measurement of cotton defects in ginned cotton within a complex background. The edge of cotton defects are extracted from fusion of three channel image of color image. A criterion based on areas is proposed to achieve fast morphological analysis. The different defects can be inspected automatically based on different thresholds. The comparison experiments between measuring system and technician were done and analyzed. The costing time of measuring system was less than 30 seconds, and accuracy was 89.5%. The measuring results show the method can meet with the requirement of grade determination of ginned cottons.
Train wheel tread will produce scrapes, peelings and other defects due to the friction between wheel and rail surface for its long-running process. Tread defects not only have a bad affect for the stability and security of the operation of the vehicle, but reduce the service life of the bearing and rail facilities and do harm for the safety and efficiency of rail transport. Among them tread scrapes and peelings are the two main defects of train tread. In order to achieve the detection and classification of tread scrapes and peelings, a method based on image processing and BP Neural Networks model was presented for detection and classification of scrapes and peelings in train wheel tread. First we preprocess the acquired images, and extract the defects. Next calculate four characteristic parameters including energy, entropy, moment of inertia and correlation, and eventually we calculate the mean and standard deviation of those characteristic parameters as the 8 texture parameters. Then we adopt principal component analysis method to turn 8 texture characteristic parameters of these two types of defects into three unrelated comprehensive variables. Finally by extracting and analysis the texture features of tread defects, the recognition correct rate reaches to 93.3%. The result shows that the method can meet the requirement of train wheel tread defects online-measurement.
The triangulation measurement is a kind of active vision measurement. The laser triangulation displacement is widely used with advantages of non-contact, high precision, high sensitivity. The measuring error will increase with the nonlinear and noise disturbance when sensors work in large distance. The paper introduces the principle of laser triangulation measurement and analyzes the measuring error and establishes the compensation error. Spot centroid is extracted with digital image processing technology to increase noise-signal ratio. Results of simulation and experiment show the method can meet requirement of large distance and high precision.
Laser collimation technology is widely applied in the positioning and measurement. The accuracy is affected by the laser beam drift, so laser beam drift compensation is necessary. Effective compensation depends on the characteristics analysis of laser beam drifts. Spectrums and values of noise signals caused by electronic noise, laser source, and environment, are analyzed in detail. The characteristics of various types of noise signals are gained and the effectiveness of low-pass filter and mean process are verified and compared. This study will provide support for separation of various types of signals and compensation of beam drifts.