Pushbroom Hyperspectral System Research Articles

The development of on-line surface defect detection system for jujubes based on hyperspectral images

• A hyperspectral system was built for collecting hyperspectral image data and performing skin defect detection of jujubes online. • Hyperspectral data of jujubes with common skin defects were obtained for training classification models. • Support vector machine and artificial neural network models were used for online skin defect detection of jujubes. • To reduce the computation time of online classification, spectral bands were selected with equal band interval method or the principal component analysis. • Experimental results showed that artificial neural network models using 14 bands can compromise the accuracy and speed for online classification of jujubes. This paper presents the development of an on-line surface defect system using hyperspectral images for jujubes. A push-broom hyperspectral system was built for collecting hyperspectral image data and detecting skin defects of jujube online. Hyperspectral images with an effective wavelength range of 468–950 nm were obtained for jujubes with normal surface or common skin defect types (i.e., russeting, decay, white fungus, black fungus and crack). Support vector machine (SVM) and artificial neural networks (ANN) models were used to classify surface defects of jujubes. The classification accuracies, with the use of full wavelength range, of ANN and SVM models for jujube skin defects are 96.5% and 96.3% respectively. The times required for processing one jujube face are about 25 and 320 s for ANN and SVM models respectively. To reduce the computation time of online classification tasks, spectral bands were selected from a wavelength range of 468–760 nm with equal band intervals or by the principal component analysis (PCA) method. Experimental results showed that the accuracy of SVM and ANN models using 14 bands (469, 491, 513, 535, 558, 580, 602, 624, 646, 668, 691, 713, 735 and 757 nm), selected by equal wavelength intervals, were 94.4% and 95% respectively. And the accuracies of ANN and SVM models with 14 bands (470, 493, 534, 555, 590, 623, 632, 654, 672, 674, 683, 696, 707 and 747) selected by PCA are 95% and 94.6% respectively. The classification time, with the use of 14 bands, of ANN and SVM models for jujube skin defects reduced to 16.6 and 30 s respectively. The online line scanning and classification hyperspectral imaging system can be used for surface defect detection of other fruits.

Evaluation of the Data Quality from a Round-Robin Test of Hyperspectral Imaging Systems.

In this study, the results from a round-robin test of hyperspectral imaging systems are presented and analyzed. Fourteen different pushbroom hyperspectral systems from eight different institutions were used to acquire spectral cubes from the visible, near infra-red and short-wave infra-red regions. Each system was used to acquire a common set of targets under their normal operating conditions with the data calibrated and processed using the standard processing pipeline for each system. The test targets consisted of a spectral wavelength standard and of a custom-made pigment panel featuring Renaissance-era pigments frequently found in paintings from that period. The quality and accuracy of the resulting data was assessed with quantitative analyses of the spectral, spatial and colorimetric accuracy of the data. The results provide a valuable insight into the accuracy, reproducibility and precision of hyperspectral imaging equipment when used under routine operating conditions. The distribution and type of error found within the data can provide useful information on the fundamental and practical limits of such equipment when used for applications such as spectral classification, change detection, colorimetry and others.

Repeatability, reproducibility, and accuracy of a novel pushbroom hyperspectral system

This study analyzes the repeatability, reproducibility and accuracy of a new hyperspectral system based on a pushbroom sensor as a means of measuring spectral features and color of materials and objects. The hyperspectral system consisted of a CCD camera, a spectrograph and an objective lens. An additional linear moving system allowed the mechanical scanning of the complete scene. A uniform overhead luminaire with daylight configuration was used to irradiate the scene using d:45 geometry. We followed the guidelines of the ASTM E2214‐08 Standard Practice for Specifying and Verifying the Performance of Color‐Measuring Instruments that define the standards and latest multidimensional procedures. The results obtained are analyzed in‐depth and compared to those recently reported by other authors for spectrophotometers and multispectral systems. It can be concluded that hyperspectral systems are reliable and can be used in the industry to perform spectral and color readings with a high spatial resolution. © 2013 Wiley Periodicals, Inc. Col Res Appl, 39, 549–558, 2014