Papers by Keyword: Auto-Focus

Abstract: This paper proposes an auto-focus method that is based on Ostu segment. The method uses the variance of the two classes of the segment results as the guidance of the auto-focus system. This guidance function responses to wide range of object distance range, it has a strong anti jamming ability and it provides a wide range of the guidance value. The result shows that this method can work well and the time it takes to focus is unconcerned to the position when the focus process starts and it responses to a wide range of object distance.

Abstract: This paper presents a design of mini actuator without using permanent magnet for auto-focus (AF) compact camera module (CCM) built on mobile phone, tablet PC and notebook. The current through the coils of electromagnets will yield magnetic force to move the moving part up and down. Comparing with conventional spring-type voice coil motor (VCM) and non-spring type VCM, the proposed actuator eliminates the concerns of rare earth supply and magnetic flux leakage which can fail the function of magnetic flux sensitive device such as e-compass. The proposed actuator also simplifies the power converter design from bipolar scheme as required by conventional non-spring type VCM to unipolar. A comprehensive electromagnetic analysis is performed by using finite element method to characterize the performance of the proposed actuator. Numerical simulations demonstrate that the magnetic flux leakage in the proposed actuator is much smaller than those by conventional design.

Abstract: The article has been discussed that it adopts step-by-step electric motor to come to auto-control through reforming the carrier and the objective lens£¬the design of the image acquisition device£¬adjustment mechanism of the Imaging lens is designed by adding the step-by-step electric motor pretending to realize high focusing accuracy and brings forward innovation design about the auto alignment device in micro field of view. This project aims to improve the ordinary microscopes and develop the digital acquisition system of the microscopic image which can accurately acquire the image information. It also can be wildly used in areas such as biomedicine, materials science, research experiments and so on.

Abstract: A new definition of potential field distribution function of image was proposed, and the characteristics of the distribution function were analyzed. Distribution function of potential field represents whole fluctuant information of image. So it can be used to analyze the image characteristics. Potential field of image background is weak and correspond to Gaussian distribution. Low stage of potential field distribution function can be used to evaluate noise level of image. Inflection point of the amplitude distribution function of potential field represented the transition point between the background and the actual edge in potential field image. So value of this point can be used as the threshold value to identify edge. When object distance of microscope changes, the edge comes more and more clear, and the edge area judged by the method above will be bigger and bigger. This character can be used to instruct auto-focus process.

Abstract: Auto-focus is a key technology for visual presenter, this paper mainly presents how to achieve the function of auto-focus on Hi3515. The 4x4 integer DCT (Discrete Cosine Transform) shown as Eq. (2) was used to indicate whether the system was on accurate focus or not. Because the 4x4 integer DCT not only has the characteristics of single peak value, non-deflection, reliability, high-speed, but also has lower complex computation than the other frequency methods such as FFT(Fast Fourier Transform), DWT (Discrete Wavelet Transform). This paper employed the method of monotonic and blind hill climbing to achieve auto-focus. The result of auto-focus is shown as Fig. 6.

Abstract: Accurate and efficient image-clarity evaluation function which adopts digital image processing technology is the key to achieving automatic focusing. The different evaluation functions are adopted for different images. By converting color images to grayscale, and comparing them with the green component which extract from the color images, it is found that using the gray image can get a much better results and an evaluation function is built which is used for tool images. Experiments prove that the used algorithm has a good single peak, accuracy, stability, and fast speed.

Abstract: The products of laser diode have been widely used, creating high profits. As there are only semi-auto-focus machines to do the focusing-jobs for the products of laser diode, it takes much work and increases the products’ cost. Therefore we developed a fully auto-focus machine for the products of laser diode to promote the quality of products and reduce the cost of the products. The AC servomotor was adopted to control the focus length. The image of the well-focused laser diode point can present more accurate information than the blurred laser point. It is helpful to get precise focus of laser diode. We adjusted the focus length of laser diode products by a fuzzy algorithm according to the feedback image information. The feedback image of laser diode spot was processed to distinguish the background, blurred region and bright region by a three-value thresholding method. The image information of the laser diode spot, the area of the bright region and the ratio between the blurred area and the bright area, were used as the criterion for the autofocusing job and to judge whether the focusing-job was well-done. In the meantime, we could also get the offsets in the x- and y- direction of the laser diode from the image of laser diode spot. Finally, we adjusted the focus length and the offsets of the laser diode products by a fuzzy imaging control algorithm to get the optimal focusing condition.

Abstract: This paper proposes a fast imaging auto-focus method to adjust the focus of laser diode products. The image of laser diode spot is processed to be a three-value image which is distinguished to be the background, blurred region and bright region. The ratio of areas between blurred and bright regions is used to judge whether the focusing is well-done. A four-point method is modified for this three-value image to get the areas of the blurred and bright regions. Thus, we can get the criterion for auto-focus. In the meantime, we can also get the offset of the laser diode from the center of laser diode spot. The four-point method is faster and more accurate than the area, three-point, contour-tracking and least-squares error methods of center-positioning. This auto-focus method of spatial domain is also faster than other methods of the energy strength in the frequency domain. Finally, we adjust the focus length of laser light by the servo control card and the AC servo motor, and then we can get the optimal focus length from the former image processing.