Jarosław Fastowicz

No-reference quality assessment of 3D prints based on the GLCM analysis

Quality assessment of 3D prints is one of the newest challenges for machine vision. As the 3D printing is relatively new technology, it is still far from perfection and there are many static or dynamically changing factors which can affect the quality of the final 3D prints. Considering the relatively long time necessary for printing, it may be reasonable to interrupt the printing process in order to save the filament (or another material) and time in the case of decreased quality of already printed fragment of an object. Such monitoring requires the use of the video feedback with appropriate image analysis methods which should allow a reliable quality assessment of the printed objects part. Typically such assessment should not be based on the comparison with a reference image, as in many image quality assessment methods, because such an image is usually unavailable. Therefore in this paper an approach for the no-reference quality assessment of the 3D prints based on the analysis of the Gray-Level Co-occurrence Matrix (GLCM) and chosen Haralick features is proposed and investigated. Obtained experimental results demonstrate the validity and usefulness of the proposed approach.

Quality Assessment of 3D Prints Based on Feature Similarity Metrics

Visual quality inspection of 3D prints is one of the most recent challenges in image quality assessment domain. One of the natural approaches to this issue seems to be the use of some existing metrics successfully applied to general image quality assessment purposes. Since the application of basic Structural Similarity does not lead to satisfactory quality prediction of 3D prints, in this paper some experimental results obtained using feature based metrics have been presented. Due to the use of different colors of filaments the influence of color to grayscale conversion method has also been analyzed. Proposed approach leads to promising results allowing a reliable prediction of 3D prints quality for different colors of filaments.

Application of Structural Similarity Based Metrics for Quality Assessment of 3D Prints

The paper is related to the verification of the usefulness of some metrics, typically used for image quality assessment and texture similarity evaluation, for no-reference quality assessment of 3D prints. The proposed approach is based on the assumption that a surface of high quality 3D print should be homogeneous and therefore some parts of it should be self-similar. Considering the local similarity of some fragments of 3D prints, some distortions can be detected which lower the overall quality of the 3D print. Since many image quality assessment methods, as well as texture similarity metrics, are based on the comparison of fragments of two images, a modification of such approach has been proposed which allows the no-reference evaluation without any information about the reference image or model.

Texture Based Quality Assessment of 3D Prints for Different Lighting Conditions

In the paper the method of “blind” quality assessment of 3D prints based on texture analysis using the GLCM and chosen Haralick features is discussed. As the proposed approach has been verified using the images obtained by scanning the 3D printed plates, some dependencies related to the transparency of filaments may be noticed. Furthermore, considering the influence of lighting conditions, some other experiments have been made using the images acquired by a camera mounted on a 3D printer. Due to the influence of lighting conditions on the obtained images in comparison to the results of scanning, some modifications of the method have also been proposed leading to promising results allowing further extensions of our approach to no-reference quality assessment of 3D prints. Achieved results confirm the usefulness of the proposed approach for live monitoring of the progress of 3D printing process and the quality of 3D prints.

Fast Machine Vision Line Detection for Mobile Robot Navigation in Dark Environments

Navigation of mobile robots based on video analysis becomes one of the most popular application areas of machine vision in automation and robotics. Recently growing popularity of Unmanned Aerial Vehicles (drones) as well as some other types of autonomous mobile robots leads to rapid increase of their application possibilities e.g. related to exploration of some areas hardly accessible for people, such as caves, underground corridors, bunkers etc. However, such places are specific in view of lighting conditions so many classical image analysis algorithms cannot be applied effectively for navigation of mobile robots in such environments. In order to utilize the image data for robot navigation in such places some modified machine vision algorithms should be applied such as fast line detection based on statistical binarization discussed in this paper.

Entropy Based Surface Quality Assessment of 3D Prints

In the paper the automatic method of visual quality assessment of surfaces of 3D prints is presented. The proposed approach is based on the use of entropy and may be applied for on-line inspection of 3D printing progress during the printing process. In case of observed decrease of the printed surface quality the emergency stop may be used allowing saving the filament, as well as possible correction of the printed object. The verification of the validity of the proposed method has been made using several prints made from different colors of the PLA filaments. Since the entropy of the image is related to the presence of structural information, the color to grayscale conversion of the test images has been applied in order to simplify further calculations. The analysis of the impact of the chosen color to grayscale conversion method on the obtained results is presented as well.

Estimation of Geometrical Deformations of 3D Prints Using Local Cross-Correlation and Monte Carlo Sampling

This paper presents a novel method for estimating the degree of deformations in 3D prints. The increased popularity of three-dimensional printing techniques introduces a necessity to create methods for quality assessment of printed materials. One of the key problems of 3D printing are strains and deformations of printed objects. This problem is determined by many factors like: printing material (filament), object geometry or temperature. The conducted research is focused on a method of automatic analysis of deformations in 3D prints based on surface scans of objects. In our research some surface scans containing varying degrees of deformations have been used for verification of obtained results. In order to evaluate the degree of deformations of materials a local cross-correlation with Monte Carlo sampling have been used. Tests carried on multiple samples have shown that the local cross-correlation technique works well when assessing the degree of deformations of printed objects on the basis of surface scans. The obtained results show that our method can be further applied for improvement of the quality of the objects received from 3D printers.

Quality Assessment of 3D Printed Surfaces in Fourier Domain.

In the paper the issue of quality assessment of 3D printed surfaces using image analysis is considered with particular attention paid to Fourier analysis of image fragments resized to one-dimensional vectors. Due to the application if Fourier analysis the regularity of visible patterns related to the consecutive layers of the filament can be assessed, assuming the side view of the printed surface. In order to avoid the problems of uneven lighting, our experiments have been conducted for scanned images of several 3D printed flat samples. As some of them have been contaminated by forced distortions, it is possible to classify them into two groups depending on the presence and amount of them. Due to the application of Fourier analysis some encouraging experimental results have been obtained which can be useful also for online monitoring of 3D prints quality for the images captured by cameras.

Color Independent Quality Assessment of 3D Printed Surfaces Based on Image Entropy

The paper is focused on the issue of visual quality assessment of 3D printed surfaces which can be helpful in detection of quality decrease during the printing process as well as the quality inspection of previously printed objects. The basic assumption of the proposed approach is the fact that each distortion of the regular patterns, visible on the side surfaces of objects printed using Fused Deposition Modeling (FDM) technology, causes the increase of the local image entropy. However, due to different colors of the filaments used in our experiments, a reliable prediction of the absolute entropy values can be troublesome. The proposed solution utilizes the combined quality indicator based on the entropy and its variance calculated for the hue component, as well as for the RGB channels, depending on the color of the filament, allowing proper detection of low quality surfaces regardless of the filament’s color.

The EOH Line Selector for Images with Downgraded Size for Mobile Robots Steering

This paper presents the idea of algorithm for detecting lines having a defined direction in digital images based on selected Region of Interest containing those lines. The proposed algorithm uses data reduction in order to simplify information processing. The goal of the algorithm is the reduction of data size based on simple random sampling method. The Edge Oriented Histogram algorithm is used to designate the ROI blocks with information of potential places in image containing lines oriented in defined direction. This approach is proposed for low-computational power systems, embedded systems or video based control of mobile robots based on image analysis. The nearly real-time algorithm has been tested on real corridor image data sets obtained from a high resolution camera. The practical test implementation as a part of mobile robot steering algorithm is presented.