Search

With the rapid development of deep learning, computer vision has assisted in solving a variety of problems in engineering construction. However, very few computer vision-based approaches have been proposed on work productivity’s evaluation. Therefore, taking a super high-rise project as a research case, using the detected object information obtained by a deep learning algorithm, a computer vision-based method for evaluating the productivity of assembling reinforcement is proposed. Firstly, a detector that can accurately distinguish various entities related to assembling reinforcement based on CenterNet is established. DLA34 is selected as the backbone. The mAP reaches 0.9682, and the speed of detecting a single image can be as low as 0.076 s. Secondly, the trained detector is used to detect the video frames, and images with detected boxes and documents with coordinates can be obtained. The position relationship between the detected work objects and detected workers is used to determine how many workers (N) have participated in the task. The time (T) to perform the process can be obtained from the change of coordinates of the work object. Finally, the productivity is evaluated according to N and T. The authors use four actual construction videos for validation, and the results show that the productivity evaluation is generally consistent with the actual conditions. The contribution of this research to construction management is twofold: On the one hand, without affecting the normal behavior of workers, a connection between construction individuals and work object is established, and the work productivity evaluation is realized. On the other hand, the proposed method has a positive effect on improving the efficiency of construction management.

The number and scale of tunnels around the world are continuously increasing, but various disease problems during the operation period have also followed, and they have become one of the important problems facing tunnels at present. Many detection methods have been proposed in the field of tunnel detection, such as traditional manual detection method, ultrasonic detection method, ground-penetrating radar method, laser scanning method and inspection method based on image processing technology. However, due to the high cost of equipment, single test content, strict test environment and other reasons, most of the current tunnel routine inspection is still manual inspection. In order to solve the existing problems in tunnel detection, a method for rapid detection and treatment analysis of cracks in tunnel linings based on deep learning is proposed. Firstly, lining cracks were selected as the main research objects, and their causes and treatment measures in different parts were analyzed. Secondly, the AlexNet convolutional neural network based on the Caffe framework was used to identify the cracks. The crack images were collected to establish a data set, and the network parameters were modified and trained. Then use MATLAB to extract the crack length and width, and design a human-machine interactive tunnel lining crack detection program in MATLAB GUI. Finally, the content and results of this paper are discussed.

In this study, a new strain measuring method was designed based on smartphone and machine vision algorithms, while the automatic calibration of the pixel size of images taken by the smartphone was achieved with the help of circle detection. A piston-type sensor to complement this method was designed, and the static and dynamic experimental results of the sensor matched well with those from FBG. Meanwhile, the sensor was tested with temperature compensation to obtain an empirical equation for the effect of temperature on the sensor. When compared with the FBG data, the static data of the sensor had the mean error of less than 4 μe and the standard error of the mean (SEM) of 1.9 μe, while the dynamic data had the mean error is less than 4.3 μe and the SEM is 5.4 μe. The results confirm that the sensor has great potential in structural health monitoring.

Currently, inclinometers are often used to monitor the horizontal displacement of deep foundation pits; however, this method generally has a high cost and complex operation and cannot monitor in real time. In this paper, a novel monitoring method for horizontal displacement of deep foundation pits based on laser image recognition technology is proposed. By identifying the displacement of the spot produced by the laser emitter fixed at the monitoring point, the displacement of the deep foundation pit is determined. In addition, the impact of the angle is considered and eliminated. A series of experimental results indicate that this method exhibits high precision and satisfies the requirements of practical engineering. Simultaneously, it exhibits the advantages of low cost, easy operation, and real-time monitoring. It provides a novel method for the displacement monitoring of deep foundation pits.

In this letter, an unsupervised method for estimating the equivalent number of looks (ENL) in synthetic aperture radar (SAR) images is proposed. Assuming that the multiplicative noise in the SAR images is a unit mean random variable of Gamma distribution parameterized by the ENL, a lookup table between the ENL and the variance of the logarithm of the multiplicative noise is established. Then, for a SAR image, we first apply the logarithmic transform to make the multiplicative noise become additive. After that, the variance of the additive noise in the logarithmic image is estimated; finally, the ENL is obtained according to the estimated noise variance by the pre-established lookup table. The effectiveness of the proposed method is demonstrated with both simulated and real SAR images.

In this letter, a censoring scheme that iteratively updates the outlier/target maps for target detection in synthetic aperture radar (SAR) images is proposed. For each iteration, any pixels that are indicated by the outlier map as outliers are rejected (censored out) from the clutter estimation. The resulting detected target map is then used as the new outlier map for the next iteration. This procedure is continued until there is no change to the target map, which is then output as the final detection result. The proposed scheme is generically applicable for target detection in both single-channel and multichannel SAR images. In our experiment, in particular, we tested the proposed method on both single-channel and polarimetric SAR data, and its effectiveness was successfully demonstrated.

Recently, the use of linear features for processing remote-sensing images has shown its importance in applications. Unfortunately, traditional linear feature detection methods rely heavily on the image's local information which makes them vulnerable to the presence of noise in the image. This problem becomes particularly difficult for synthetic aperture radar (SAR) image applications where SAR images are corrupted by speckle noise. In order to overcome this problem, we propose a novel method that processes the polarimetric synthetic aperture radar (Pol-SAR) images by combining the multiscale image analysis with polarimetric information in a new fashion. A two-scale approach is adopted here. On a coarse level, the coarse regions of the linear features are extracted by a curvelet transform from a speckle noise reduced image obtained by the polarimetric whitening filter. On a fine level, we develop a fuzzy polarimetric detector to accurately locate the linear features inside the regions. The effectiveness of the proposed method is demonstrated using simulated Pol-SAR data acquired from both EMISAR and Convair-580 systems.

Visual interpretation of SAR images plays a critical role in remote sensing applications. For rapidly obtaining the image suitable for human observation, a new method of visualization for SAR images is proposed in this paper. The statistics of the SAR data, including the single channel data intensity and the span of the Pol-SAR data, are derived. The proposed method is a parameterized method based on the characteristic statistic of the SAR data. Using the AIRSAR data, the effectiveness of the proposed method is demonstrated by comparing to the traditional algorithm.

In this paper, a new method for supervised classification of terrain types in polarimetric Synthetic Aperture Radar (Pol-SAR) images is proposed. This technique is a combination of the texture classification and the maximum likelihood classification based on the complex Wishart distribution for the polarimetric covariance matrix. The texture features are first extracted from the span image based on co-occurrence matrices; and then the classifier combines the texture features with the distance measure based on polarimetric information to obtain the results. Using a NASA/JPL AIRSAR image, the effectiveness of the proposed method is demonstrated.

A new method for edge detection of polarimetric SAR images is proposed based on the curvelet transform and the Duda operator in this paper. The detection of edge features from SAR images is complicated by the presence of speckle. Firstly, a despeckled polarimetric SAR image is derived by the Polarimetric Whitening Filter (PWF), and then the image is filtered by Lee's method with eight edge-aligned windows to preserve the edge sharpness. The curvelet transform which is a multi-scale representation suited for edge enhancement is applied to estimate the area including edges approximately. Then the Duda operator, a contrast ratio edge detector to extract linear features, is used to give the positions of the edges precisely. The proposed method is applied to PISAR images to illustrate its effectiveness. (4 pages)