Visual Inspection System Based On The Gray Scale Morphology

This paper presents a new method of dorsal hand vein images segmentation based on local thresholding using grayscale morphology. The images, captured by near- infrared CMOS cameras, have poor contrast, speckling noise and non-uniform lighting. After the preprocessing these images normalized, we use dilation and erosion operations to calculate the local thresholding and segment to get the binary images. We do experiments on 40 images effectively. Hence, the research will be significant for real-time applications using infrared vein image segmentation, such as identification systems and vein location or view for humans.

One of the challenges faced by the users of automated visual inspection (AVI) systems is how to efficiently upgrade the legacy systems to inspect the new components introduced into the assembly lines. If the AVI systems are not flexible enough to accommodate new components, they will be rendered obsolete even by small changes in the product being inspected. The overall objective of the research presented in this paper is to produce the methodological basis that will result in the development of highly reconfigurable AVI systems. In this paper, we focus on part of this overall development, the adaptation of preexisting inspection algorithms to inspect similar components introduced into the assembly line. While this paper bases its development and discussion on the inspection of surface mounted devices (SMDs), the proposed methodology is general enough to be applicable to a broad range of inspection problems. In this paper, we present a methodology that would allow the automation of the refinement of AVI algorithms. In particular, the proposed method identifies a set of components, or cluster of components, for which a particular set of inspection features or algorithms, renders a certain level of inspection reliability. This is particularly useful for adapting preexisting systems to inspect new components, especially when the characteristics of the new components are similar to those of components already inspected by the inspection system. We applied this methodology to a case of study of the inspection of SMDs.

Visual quality enhancement plays a vital role in low cost imaging systems, machine vision, industrial applications, remote sensing, face recognition systems and medical image interpretation etc. Growth of low cost image processing applications require image preprocessing which enhances details of an image. Most of the contrast enhancement papers apply desired contrast enhancement technique directly to enhance the given input image having poor contrast or contrast at any other undesired level. It is important to predict whether the contrast enhancement is needed for an image, to avoid the artifacts due to enhancement on the good image. In this paper, an algorithm to model images using its local contrast measure has been proposed, to classify and distinguish between the images having different contrast level. The input image is classified either as low contrast or high contrast image using the model. If the classified image is low contrast it will be enhanced using the Stochastic Resonance principle. The results show that the proposed automated procedure enhances the low contrast image better than the conventional enhancement methods.

A rule-based system must satisfy stringent timing constraints when applied to a real-time environment. As the scale of rule-based expert systems increases, the efficiency of systems becomes a pressing concern. The most critical performance factor in the implementation of a production system is the condition-testing algorithm. We propose a new method based on the widely used RETE match algorithm. We show an approach designed to reduce the response time of rule-based expert systems by reducing the matching time. There are two steps in the method we propose: The first makes an index structure of the tokens to reduce the /spl alpha/-node-level join candidates. The second chooses the highest time tag for certain /spl beta/-nodes to reduce the amount of combinatorial match that is problematical in a real-time production system application. For this purpose, a simple compiler is implemented in C and the response time of test programs is measured.

The premise that one can obtain more effective control systems with less engineering effort, less technical risk, and less cost via the use of artificial intelligence techniques is explored in this paper. In particular, we discuss, develop, and illustrate the use of a rule-based system for real-time control as a software environment that can be used for control system implementation, relieving the design engineer of much of the burden of taking a basic control algorithm and making it work in a real-world application. We show that the rule-based systems approach can be used to provide substantial support for the above task. A rule-based system provides the environment in which the designer can develop and test all of the required heuristic logic and control, using a programming language (production rules) that is ideally suited for the task. We develop the details of such an implementation methodology, and illustrate its application to failure detection and isolation.

This paper presents the fuzzy image enhancement for low contrast and non-uniform illumination images. A new fuzzy intensity measure is proposed to distinguish between the dark and bright regions. This measure is computed by considering the average intensity and deviation of the intensity distribution of the image. The input image is enhanced using a power-law transformation. Implementation of the proposed algorithm on the non-uniform illumination and low contrast images show that the proposed algorithm outperforms the other enhancement techniques. The proposed algorithm produces more even illumination with improvement in the details and contrasts. In addition, the proposed algorithm is computationally fast to be implemented in real time application.

Salient object detection has drawn much attention recently. The extracted salient objects can be used for tasks including recognition, segmentation and retrieval. Most existing models for saliency detection can only be applied to the high contrast and normal contrast scenes. They are not robust enough to handle objects in low contrast images. Due to low lightness, low Signal to Noise Ratio (SNR) and low appearance information, saliency detection in low contrast scenes remains a challenge. To solve this problem, this paper proposes a saliency detection model based on the feature extraction, in which optimal features have been learned in different contrast environment to improve the effectiveness of salient object detection. Compared with other existing saliency models, the proposed method has strong adaptability for different scenes, and gives superior performance in detecting the salient object especially in low contrast images.

Due to the variety of types of images, there are currently no universal methods that provide a guaranteed result of solving the problem of improving the quality of low-contrast digital images. Medical systems often produce images of insufficient quality for reliable visual analysis. In particular, X-ray images, characterized by low intensity, un-even background, high level of noise, poor contrast and weakly defined boundaries of structures, are particularly difficult to analyze and choose an effective processing meth-od. The paper presents the information possibilities of the method of processing half-tone medical images aimed at improving the contrast and increasing the detail of ob-jects of interest in order to increase the reliability of diagnosis based on them. The pro-posed algorithm is based on a multi-stage processing process, which includes the use of two-dimensional frequency Fourier transformation and the method of fuzzy intensifica-tion in the spatial domain. The use of two-dimensional discrete Fourier transformation changes not the im-age, but the form of its representation, converting the output signal into its components of different frequencies and amplitudes. In this form, it is much easier to carry out filter-ing or amplification of individual components of the signal. Fuzzy set theory has the ability to quantitatively and qualitatively model problems related to uncertainty and imprecision, which are always present in digital images. Their presence is determined both by the features of the physical processes of image forming systems and by the stage of creating a digital image. The application of the proposed method provides improved contrast and increased detailing of objects of interest and affects the reliability of visual analysis. Only the ar-guments of complex Fourier transform coefficients can be corrected. The method of fuzzy intensification is used as a refinement for the second stage of frequency conversion. The effect of frequency conversion parameters on the detail of the resulting image was stud-ied. The results of the algorithm are presented on the example of real X-ray images.

21 - Automated Visual Inspection Systems

In the field of image processing, improving low-contrast images is essential for enhancing the visibility of features that may be hard to see due to slight intensity variations. This issue often arises with images captured in poor lighting or unfavorable conditions, which can impede tasks such as feature extraction, object recognition, and edge detection. This paper presents an algorithm that employs fuzzy techniques to enhance low-contrast images. The approach utilizes a Pythagorean fuzzy set, which is an advanced version of a fuzzy set. The Pythagorean fuzzy set, by integrating membership and non−membership degrees, offers a more precise representation of image uncertainty. The proposed method constructs an increasing function to calculate the degree of non−membership, thereby amplifying the effectiveness of the enhanced image. By integrating this method with contrast limited adaptive histogram equalization, alongside a color restoration technique and brightness-preserving dynamic histogram equalization, the brightness and natural appearance of the original low-contrast image are maintained. We compare our approach to other methods, including conventional histogram equalization, intuitionistic fuzzy generators, interval−valued intuitionistic fuzzy generators, interval−valued intuitionistic fuzzy with contrast limited adaptive histogram equalization, LightNet, and FlightNet. Experimental results indicate that our method surpasses the techniques currently utilized in terms of visibility and color accuracy. Performance evaluation is performed using metrics such as the absolute mean brightness error, contrast improvement index, and structural similarity index.

Face recognition is used to identify one or more persons from still images or a video image sequence of a scene by comparing input images with faces stored in a database. The face images used for matching the image in the database has to be of good quality with normal lighting condition and contrast. However, face images of poor illumination or low contrast could not be recognized properly. The objective of the work is to enhance the facial features eyes, nose, and mouth for poor contrast facial images for face recognition. The image enhancement is done by first detecting the face part, then applying contrast-limited adaptive histogram equalization technique and thresholding to enhance the facial features. The brightness of the facial features is enhanced by using logarithm transformation. The proposed image enhancement method is implemented on AR database, and the face images appear visually good when compared to original image. The effectiveness of the enhancement method is compared by analyzing the histogram.

Accurate segmentation of anatomical brain structures is crucial for many neuroimaging applications, e.g., early brain development studies and the study of imaging biomarkers of neurodegenerative diseases. Although multi-atlas segmentation (MAS) has achieved many successes in the medical imaging area, this approach encounters limitations in segmenting anatomical structures associated with poor image contrast. To address this issue, we propose a new MAS method that uses a hypergraph learning framework to model the complex subject-within and subject-to-atlas image voxel relationships and propagate the label on the atlas image to the target subject image. To alleviate the low-image contrast issue, we propose two strategies equipped with our hypergraph learning framework. First, we use a hierarchical strategy that exploits high-level context features for hypergraph construction. Because the context features are computed on the tentatively estimated probability maps, we can ultimately turn the hypergraph learning into a hierarchical model. Second, instead of only propagating the labels from the atlas images to the target subject image, we use a dynamic label propagation strategy that can gradually use increasing reliably identified labels from the subject image to aid in predicting the labels on the difficult-to-label subject image voxels. Compared with the state-of-the-art label fusion methods, our results show that the hierarchical hypergraph learning framework can substantially improve the robustness and accuracy in the segmentation of anatomical brain structures with low image contrast from magnetic resonance (MR) images.

In order to increase the performance of automatic sleep stage scoring, we propose a hybrid neural-network and expert system taking advantages of each system. After signal cleaning and feature extraction from polysomnographic signals using several algorithms we suggested, the rule-based expert system classified the sleep states with symbolic reasoning. The neural network supplemented the shortcomings of rule-based system by dealing with exceptions of rules. The result shows that the combination of computational and symbolic intelligence is promising approach to automatic sleep signal analysis.

Land use suitability assessment for economic development at the provincial level: The case study of Yazd Province, Iran

The Histogram of Oriented Gradients (HOG) is a popular feature descriptor used in computer vision and image processing. The technique counts occurrences of gradient orientation in localized portions of an image. The descriptor is sensible to the presence in images of noise, nonuniform illumination, and low contrast. In this work, we propose a robust HOG-based descriptor using the local energy model and phase congruency approach. Computer simulation results are presented for recognition of objects in images affected by additive noise, nonuniform illumination, and geometric distortions using the proposed and conventional HOG descriptors.