Accurate modeling and prediction of surface roughness by computer vision in turning operations using an adaptive neuro-fuzzy inference system

doi:10.1016/S0890-6955(02)00078-0

International Journal of Machine Tools and Manufacture

Volume 42, Issue 13, October 2002, Pages 1441-1446

https://doi.org/10.1016/S0890-6955(02)00078-0 Get rights and content

Abstract

Modeling and prediction of surface roughness of a workpiece by computer vision in turning operations play an important role in the manufacturing industry. This paper proposes a method using an adaptive neuro-fuzzy inference system (ANFIS) to accurately establish the relationship between the features of surface image and the actual surface roughness, and consequently can effectively predict surface roughness using cutting parameters (cutting speed, feed rate, and depth of cut) and gray level of the surface image. Experimental results show that the proposed ANFIS-based method outperforms the existing polynomial network-based method in terms of modeling and prediction accuracy.

Introduction

Standard roughness measurement procedures depend heavily on stylus instruments which have only limited flexibility in handling different parts. Furthermore, the procedure is a post-process approach, which is not amenable for automation, and the measurement is also relatively slow. In recent years, the modeling and prediction problems of surface roughness of a workpiece by computer vision in turning operations have received a great deal of attention [1], [2], [3], [4], [5], [6]. Although it has been shown that the surface roughness is strongly characterized by the surface image, practical surface roughness measurements based on computer vision technology are still difficult [1]. The main problem is how to accurately obtain the actual surface roughness of a workpiece using surface images and various parameters of cutting operations. Lee and Tarng [1] used a self-organizing adaptive learning tool called polynomial network (PN) [7] to establish the relationships between the actual surface roughness of a workpiece and the feature of surface image under a variety of cutting operations. As a result, the PN-based method can effectively predict the surface roughness with reasonable accuracy. The reported mean error of prediction of surface roughness is about 6.2%.

To improve the accuracy of predicting surface roughness, this study uses a more powerful learning tool called fuzzy neural network (FNN) [8], [9], [10] to solve the problems of modeling and prediction of surface roughness. It is known that the adaptive neuro-fuzzy inference system (ANFIS) is efficient for non-linear mapping. ANFIS is a fuzzy inference system implemented in the framework of an adaptive FNN. By using a hybrid learning procedure, ANFIS can construct an input–output mapping based on both human knowledge (in the form of fuzzy if–then rules) and stipulated input–output data pairs [8], [9]. In this paper, we utilize ANFIS to model the relationship between the features of surface image and the actual surface roughness. Consequently, once the cutting parameters (cutting speed, feed rate, and depth of cut) and gray level of the surface image are given, surface roughness can be accurately predicted. Using the same group of surface images and turning parameters as that in [1], the average errors of modeling and prediction are 5.88×10⁻⁸% and 0.38%, respectively. The encouraging experimental results show that the proposed ANFIS-based method outperforms the existing PN-based method [1] in terms of modeling and prediction accuracy.

Section snippets

Computer vision system for inspecting surface roughness

For comparison with the PN-based method [1], the same group of surface images and turning parameters is used in this study. The computer vision system for inspecting surface roughness adopted by Lee and Tarng [1] is briefly described as follows. The schematic diagram of the computer vision system including a vision system with a digital camera (Olympus C-1400L) and an appropriate lighting arrangement is shown in Fig. 1. The camera captures surface images with 1280 × 1024 resolution, 1/30 s

The proposed ANFIS-based method

The architecture of the ANFIS used in the proposed method is shown in Fig. 2. There are four input parameters (V, F, D, Ga) and one output value, the predicted surface roughness $R a$ . Denote the output node i in layer l as O_l,i. The used five-layer ANFIS is described as follows:

Layer 1: Every node in this layer is an adaptive node with a node output defined: O_1,i=μ_{v_i}(V), for i = l,…, m, O_1,i+m=μ_{F_i}(F), for i = l,…, m, O_l,i+2m=μ_{D_i}(D), for i = l,…, m, O_1,i+3m=μ_{Ga_i}(Ga), for i = l,…, m, where V (or F,

Modeling and prediction of surface roughness

The experimental data of 57 turning experiments listed in Table 1 are utilized to train the used ANFIS model with m = 3. The hybrid batch learning rules are used in the training [8], [9], [10]. The values of premise and consequent parameters obtained after training are given in the Appendix. The trained ANFIS establishes the relationship between the features of surface image and the actual surface roughness. Once the cutting parameters (V,F,D) and the gray level Ga are given, then the predicted

Conclusion

This paper proposes a method using an adaptive neuro-fuzzy inference system (ANFIS) to accurately establish the relationship between the features of surface image and the actual surface roughness, and consequently can effectively predict surface roughness using cutting parameters (cutting speed, feed rate, and depth of cut) and gray level of the surface image. The advantages of the proposed method are non-contact measurements, ease of automation, and high accuracy. Experimental results have

References (10)

B.Y Lee et al.
Surface roughness inspection by computer vision in turning operations
International Journal of Machine Tools and Manufacture
(2001)
S Damodarasamy et al.
Texture analysis using computer vision
Computers in Industry
(1991)
D.E.P Hoy et al.
Surface quality assessment using computer vision methods
Journal of Materials Processing Technology
(1991)
G Galante et al.
Surface roughness detection by tool image processing
Wear
(1991)
M.B Kiran
B Ramamoorthy, V. Radhakrishnan, Evaluation of surface roughness by vision system
International Journal of Machine Tools and Manufacture
(1998)

There are more references available in the full text version of this article.

Cited by (128)

Machine vision based condition monitoring and fault diagnosis of machine tools using information from machined surface texture: A review
2022, Mechanical Systems and Signal Processing
Machine vision based condition monitoring and fault diagnosis of machine tools (MVCMFD-MTs) is a vital technique of condition-based maintenance (CBM) in both metal removal manufacturing and metal additive fabrication. In these domains, many methods utilize information from imaging matrices of machined surfaces to extract sensitive features and obtain potential degradation tendencies. Over recent years, no comprehensive review covers the whole monitoring or diagnostic procedures. To fill this gap, this paper systematically summarizes MVCMFD-MTs, which aims to provide researchers and engineers with a theoretical basis and roadmap to further study or build MVCMFD-MTs using information from the machined surface texture. Firstly, two data acquisition systems and several institutional public datasets are revisited. Secondly, the methodologies are illustrated in two aspects, feature descriptors and diagnostic decision-making. Thirdly, an intuitive illustration on applications is provided from the perspective of surface quality monitoring (i.e., roughness evaluation, surface defect inspection) and indirect tool condition monitoring (i.e., tool wear monitoring, chatter identification). Finally, this paper discusses current challenges and potential research directions in nowadays intelligent manufacturing.
An approach for surface roughness measurement of helical gears based on image segmentation of region of interest
2021, Measurement: Journal of the International Measurement Confederation
Existing roughness measurement approaches based on machine vision cannot accurately measure irregular components with complex shapes, such as helical gears. Owing to the occlusion of relative positions between teeth, it is not possible to directly obtain an image that only contains the target surface, which decreases the accuracy and efficiency of the measurement model. This paper proposes a novel visual approach for the roughness measurement of helical gears. First, a region of interest (ROI) extraction method is designed to filter the interference information in the original image and extract the effective region. Then, a convolutional neural network (CNN) is applied to evaluate the roughness with the ROI processed image as input. The machine vision-based roughness values calculated before and after ROI extraction are compared with the stylus device-based roughness values. The accuracy and generality of the proposed approach are proved by two cases of helical gear and leadscrew roughness measurements.
Modeling of friction stir welding process using adaptive neuro-fuzzy inference system integrated with harris hawks optimizer
2019, Journal of Materials Research and Technology
Friction Stir Welding (FSW) has been paid more attention in recent years due to its efficiency in welding materials that are difficult to weld by conventional fusion welding methods. There are several parameters that affect FSW process, so it is important to understand the relationship between different process parameters to maximize the quality and strength of the joint. This paper proposed an alternative method to predict FSW parameters and make a decision using a modified version of the adaptive neuro-fuzzy inference system (ANFIS) integrated with harris hawks optimizer (HHO). HHO was used to search for optimal values of ANFIS parameters and to determine the optimal operating conditions of the FSW process. The shared effect of welding speed, tool rotational speed, and plunge force on the mechanical properties of welded aluminium plates was simulated. The proposed model, called ANFIS-HHO, was used to predict the mechanical properties of FSW Al plates in terms of ultimate tensile strength (UTS) as functions of welding speed, tool rotational speed, and plunge force. The adequacy of the model was tested; the predicted data were in good agreement with the experimental data. The tool rotational speed and the empirical force index (EFI) have a significant impact on the mechanical properties of the welded joints. ANFIS-HHO technique was found to be a powerful optimization tool for predicting FSW parameters to achieve high joint strength.
Illumination Compensated images for surface roughness evaluation using machine vision in grinding process
2019, undefined
Appropriate lighting is one of the indispensable elements in inspection using machine vision system. Illumination variation affects the accuracy and robustness of an inspection method that employs a machine vision system. The lighting inhomogeneity is the disturbing signal that needed to be suppressed to achieve accuracy and consistency in surface roughness quantification. In this work, the illumination compensation techniques are used for ground surface roughness evaluation by statistical texture parameters using machine vision method. The three-dimensional (3-D) surface roughness parameters are compared with the texture parameters. The experimental results are based on the ground surface images that are machined at different machining parameters. After the grinding process, the images are captured under halogen lighting. The acquired images of ground specimens are used for illumination compensation using: homomorphic filtering, Discrete Cosine Transform (DCT) based filtering and Fourier Transform (FT) based filtering techniques. This helps to suppress the low frequency components and amplify the high frequency components in order to extract the texture information. Owing the fact that the ground surfaces were weaker anisotropic surfaces, the second order statistical evaluation methods are used to extract the changes in the image texture due to the variation in surface roughness of the component. The texture parameters evaluated using these methods are correlated with the 3-D surface roughness parameters measured using an optical profiler. The texture parameters showed better correlation with the measured roughness values and this can be an integral part of any grinding system to inspect the machined components. In order to establish the homogeneity achieved after compensation of images, the inhomogeneity indicator and harmonic distortion values are calculated for the ground images.
A fuzzy multi-criteria spatial decision support system for solar farm location planning
2017, Energy Strategy Reviews
In recent years, investment in solar energy has increased substantially across countries. Thus, selecting convenient locations for solar farms has become a fundamental problem when determining the investment required due to differences in climatic factors, the type and availability of land, transportation infrastructures, and the quality of power lines. Multi-Criteria Evaluation approaches based on crisp data are generally used in the selection process of optimal locations. However, despite being crisp, the data available when considering the evaluation criteria of the different alternatives constitute a discrete approximation performed on a spatial grid of potential locations. Thus, we introduce a three-stage fuzzy evaluation framework designed to account for the imprecision inherent to the evaluations when identifying the most convenient location for constructing solar power farms. First, we implement ANFIS (Adaptive Neuro-Fuzzy Inference System) on the set of grid intersection crisp data points and derive a coherent set of approximations per each potential discrete location and evaluation criterion. Then, the fuzzy AHP (Analytic Hierarchy Process) method is used to determine the weights of the different criteria considered from the linguistic evaluations provided by different experts. Finally, we define a set of if-then rules combining the different ANFIS evaluation criteria and their weights within a FIS (Fuzzy Inference System) whose output is used to determine the most convenient location for constructing a solar power farm. The efficacy of the proposed evaluation framework is demonstrated through its application to the Iranian regions of Kerman and Yazd.
Computer vision in manufacturing: a bibliometric analysis and future research propositions
2023, International Journal of Advanced Manufacturing Technology

View all citing articles on Scopus

View full text

Accurate modeling and prediction of surface roughness by computer vision in turning operations using an adaptive neuro-fuzzy inference system

Abstract

Introduction

Section snippets

Computer vision system for inspecting surface roughness

The proposed ANFIS-based method

Modeling and prediction of surface roughness

Conclusion

International Journal of Machine Tools and Manufacture

Computers in Industry

Journal of Materials Processing Technology

Wear

International Journal of Machine Tools and Manufacture