Specular highlight region restoration using image clustering and inpainting

Kang, Ho-Sun; Hwang, Dokyung; Lee, Jang-Myung

doi:10.1016/j.jvcir.2021.103106

Cited by 5 publications

(5 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Its calculation formula is shown in (17). Evaluation results in Table 1 show that the proposed method not only has good detection accuracy but also has the least detection time compared with those in [11][12][13]24]. The average detection time is about 0.64 s, which has good real-time performance.…”

Section: Reflective Area Detectionmentioning

confidence: 98%

“…In this figure, the white parts are identified as the reflective areas of the images detected. It can be seen from Figure 18 that the detection results of the proposed method and those in [13] are good, indicating that the reflective area can be accurately detected, while there are a few false detections by using that in [24]. Additionally, the detection results by using those in [11,12] are poor, including obvious false detections.…”

Section: Reflective Area Detectionmentioning

confidence: 99%

“…The improved k-means algorithm is used to detect the reflective area and compared with other detection methods available in the literature. Figure 18 shows the detection results of the proposed detection methods and those in [11][12][13]24]. In this figure, the white parts are identified as the reflective areas of the images detected.…”

Section: Reflective Area Detectionmentioning

confidence: 99%

“…The evaluation results of the average detection time and detection accuracy, i.e., the intersection over Union (IoU) of the reflective area, are listed in Table 1. Those methods in [11][12][13] and [24] in the same experimental conditions are also listed in Table 1. The IoU here is an evaluation index characterizing the accuracy of the marked reflective area.…”

Section: Reflective Area Detectionmentioning

confidence: 99%

“…Cao et al [12] determined the reflective area by solving the non-mirror reflectance estimation map and then solving the diffuse reflectance chromaticity estimation of each pixel. Kang et al [13] converted images to the HSI model and filtered the reflective pixels by the combination of the k-means clustering algorithm and filter. However, these methods have limited applications due to their shortcomings in precision, speed, and complex illumination conditions.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Adaptive Reflection Detection and Control Strategy of Pointer Meters Based on YOLOv5s

Liu

Deng

Zhang

et al. 2023

Sensors

View full text Add to dashboard Cite

Reflective phenomena often occur in the detecting process of pointer meters by inspection robots in complex environments, which can cause the failure of pointer meter readings. In this paper, an improved k-means clustering method for adaptive detection of pointer meter reflective areas and a robot pose control strategy to remove reflective areas are proposed based on deep learning. It mainly includes three steps: (1) YOLOv5s (You Only Look Once v5-small) deep learning network is used for real-time detection of pointer meters. The detected reflective pointer meters are preprocessed by using a perspective transformation. Then, the detection results and deep learning algorithm are combined with the perspective transformation. (2) Based on YUV (luminance-bandwidth-chrominance) color spatial information of collected pointer meter images, the fitting curve of the brightness component histogram and its peak and valley information is obtained. Then, the k-means algorithm is improved based on this information to adaptively determine its optimal clustering number and its initial clustering center. In addition, the reflection detection of pointer meter images is carried out based on the improved k-means clustering algorithm. (3) The robot pose control strategy, including its moving direction and distance, can be determined to eliminate the reflective areas. Finally, an inspection robot detection platform is built for experimental study on the performance of the proposed detection method. Experimental results show that the proposed method not only has good detection accuracy that achieves 0.809 but also has the shortest detection time, which is only 0.6392 s compared with other methods available in the literature. The main contribution of this paper is to provide a theoretical and technical reference to avoid circumferential reflection for inspection robots. It can adaptively and accurately detect reflective areas of pointer meters and can quickly remove them by controlling the movement of inspection robots. The proposed detection method has the potential application to realize real-time reflection detection and recognition of pointer meters for inspection robots in complex environments.

show abstract

Section: Reflective Area Detectionmentioning

confidence: 98%

Section: Reflective Area Detectionmentioning

confidence: 99%

Section: Reflective Area Detectionmentioning

confidence: 99%

Section: Reflective Area Detectionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations