Papers by Keyword: Machine Vision

Abstract: In order to improve measurement accuracy and reduce maintenance work of dynamic measurement system, research is proposed on non-contract measurement method based on machine vision. To choose real-time measurement system of mining car as test object, and use binocular vision to calculate 3 D coordinates of contour, and obtain the values of volume and weight with material through operation. Test platform and work flow is designed in detail. A calibration method combined the advantages of optical calibration with camera calibration itself is proposed. To simplify the extraction and matching of feature points through using structured light. Test results show that this method can effectively improve the accuracy of dynamic measurement system for bulk material.

Abstract: This paper presents an novel vision based inspector with light to inspect the liquid in bottle. The inspector is designed to capture sequence images of the rotating liquid in the bottle. And the impurities in liquid are inspected by dynamic analysis. The difference method for fusion images is put forward to detect the motion regions in sequence images. In order to detect motion regions precisely, this paper employs a novel segmentation algorithm base on unsupervised learning. This algorithm combines the fuzzy C-means method with fuzzy support vector machines, and can subdivide the image into the impurities and background efficiently. The experiments demonstrate the inspection precision of the liquid inspection system is about 96.4%.

Abstract: According to the characteristics of infrared imaging and rare bird species, a infrared imaging inspection method for rare bird species based on nonlinear filter algorithm has been proposed. Firstly, the CCD(charge-coupled device) sensors are used to obtain infrared video-data for rare birds, and then the captured images of rare birds are preprocessed to obtain the body axis length. Secondly, nonlinear filter algorithm, combined with physiological characteristics of rare bird species, is used to detect and identify the high-speed flying birds. Finally, the capabilities of EKF(extended Kalman filter) and PF(particle filter), the two kinds of nonlinear filter algorithm in inspecting have been researched deeply. Theoretical analysis and experiments demonstrate that the performance of PF is much better than EKF to inspect high-speed flying non-rigid object in complex environment.

Abstract: For the reason of difficult to get battery box pose information, we research the battery box pose measure method based on visual information. We get the coplanar four points at the lines constraints which extracted form image. We get the pose relationship between the battery box coordinate system and camera coordinate system, and then calculate the average of the measure results to reduce noise effects for measure precision. Simulation results show that the method through calculate average of the measure results can effectively reduce noise effects for measure accuracy. The actual experimental results show that the pose estimate accuracy is meet robot requirements for battery swap.

Abstract: On-line detection method for automobile frame side rail process holes is proposed in this articled. It is achieved by virtue of machine vision technology detection method. Many images captured by CCD camera are processed and analyzed to finally complete the automatic detection of automobile chassis frame process holes. Machine vision technology is applied to achieve the on-line detection of machining quality of frame side rail mounting holes. The developed detection system prototype has very high detection accuracy.

Abstract: The study was conducted to identify three types of non-touching grain kernels using a colour machine vision system. Images of individual cereal grain kernels were acquired using an camera. Shape feature was extracted from binary and edge images of cereal grain kernels obtained by iamge processing for classification. A total of 13 shape feature parameters, including region area, perimeter, length, width, the maximum radius, the smallest radius etc, were extracted from each kernel to use as input to the Bayesian classifier. Experimental results showed that the Bayesian classifier gave better classification with a calssificaiton accuracy of 99.67% for indica type rice, followed by 98.67% and 78.33% for japonica rice and glutinous rice using training set, respectively. The classification system was developed with Bayesian classifier that achieved an overall recognition rate of 92.22% with training data set and furthermore, a classification accuracy of 90% for the testing data set.

Abstract: A new method for saliency detection is presented. Based on the sparse coding model, we propose a power spectral filter to eliminate the second-order residual correlation, which suppress the global repeated items effectively. In addition, aim to modeling the mechanism of the human retina prior response to high-contrast stimuli, the effect of color context is considered. Experimental result indicates that our method has high-quality detection performance with respect to the ability not only to highlight the salient objects in complex environment but also to pop up them uniformly.

Abstract: This paper presents an image detection algorithm for navigation route of cotton harvester. Two cameras were respectively installed on the leftmost and rightmost picker unit, and images were captured during working process respectively. Firstly, the color characteristics among harvested field, un-harvested field, outside-field and the end of field were analyzed, then the target features of different fields was extracted using the color difference 3B-R-G. Secondly, candidate point group was determined by looking for the critical point of peak from the lowest trough point to un-harvested field and associating with the detection result of the anterior frame. Lastly, navigation line was obtained by using passing a Known Point Hough Transform (PKPHT). Results show that the navigation line detected using this algorithm can fit the boundary line and the edge of field accurately, the average processing time is56.10ms/f, and the algorithm can meet the actual production needs of cotton harvester.

Abstract: In order to realize automatic operation for a wheat planter in a field, an algorithm was developed in the research to detect navigation line under sowing operating environment with weak navigation information based on machine vision. Wavelet transform, linear analysis and front and rear frame interrelated were used to get candidate points at regional boundary in the image. Then linear fitting of the candidate points was carried out using the Passing a Known Point Hough Transform. Sowing videos captured under different natural conditions, in different regions were used to test the performance of the algorithm. Results show that the algorithm is able to detect ridge line, sowing line and field end accurately, steadily and quickly, the average processing time for each frame is about 30ms.

Abstract: A high efficient method is provided to count the number of kernels in an ear corn. PC camera captured a sequence of images around an ear corn using a simple device. The Otsu's algorithm was applied for background segmentation. An object processing area was obtained after contour extraction. The x-direction cumulative histogram was used to extract every row of the ear corn. The y-direction cumulative histogram was used to detect the number of corn kernels in a row. The number of rows was got by matching the edge of the current ear row with the first one. The time of detecting kernels for an ear corn was below 2 minutes and the average accuracy was about 95% in experiment. It shows that this method can detect the quantity of kernels directly in an ear corn quickly and effectively using a low-cost device.