Comparison Between Traditional and CNN Based Stereo Matching Algorithms

Kim, Jong Min; Park, Jeong-Min; Lee, Joon-Woong

doi:10.5302/j.icros.2020.20.0032

Cited by 3 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our experiments, we evaluate an algorithm using two criteria: matching Success Rate and algorithm runtime. We first define the matching accuracy, which is defined in (14).…”

Section: Experimental Results and Analysismentioning

confidence: 99%

“…At present, most multisource image template matching algorithms achieve template matching by extracting structural features from the grayscale image and then evaluating the similarity between dense structural features [13], [14], [15], [16]. Recently, Lu et al [13] proposed a template matching algorithm based on Structure Tensor Voting and Orientation (STVO), which is a dense feature descriptor algorithm but has a much higher computing efficiency.…”

Section: Related Workmentioning

confidence: 99%

“…Existing template matching methods can be roughly divided into the traditional method and the deep learning method based on neural networks. Traditional methods usually measure the similarity between two images through the statistical difference of grayscale information or the distance between dense feature vectors [13], [14], [15], [16], such as Normalized Cross Correlation (NCC) [17], [18], [19], Sum of Squared Difference (SSD) [20], [21], Mutual Information (MI) [22], Local Self-Similarity (LSS) [23], [24], etc. The pixel-level grayscale information statistics and complex dense feature extraction lead to the inefficiency of this type of algorithms in the early stage.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Fast Template Matching Scheme of Visible and Infrared Image Under Occluded Scenarios

Mei

Wang²,

Wang

et al. 2022

IEEE Access

View full text Add to dashboard Cite

A fast and robust template matching scheme, called Matching by Slice Transform Matrix Mapping (MSTMM), is proposed for the matching difficulty under occluded scenarios caused by nonlinear intensity differences and structure differences between visible and thermal infrared images. The first step in the MSTMM scheme was to extract information about the distribution of pixels with the same gray level through the developed Expanded Slice Transform with Adaptive Gray Level (EST-AGL). After completing the construction of the EST-AGL matrix for all image patches, different EST-AGL matrices were mapped to different integers or floats through the traditional special integer mapping mechanism or the neural network mapping mechanism. Finally, template matching between visible and thermal infrared images was achieved by evaluating the similarity of correlation mapping surface images through the Normalized Cross Correlation (NCC) algorithm. The proposed EST-AGL method can overcome the nonlinear intensity differences between visible and thermal infrared images by extracting the structural features of the image. The mapping mechanism of the MSTMM scheme can reduce the structural differences between the normal template image and the query image under an occluded scenario by increasing the similarity between the normal image patches and the image patches with occlusion. The proper mapping mechanism ensures the high performance of the MSTMM scheme by using only the simple NCC algorithm instead of other time-consuming anti-occlusion dense feature algorithms in the similarity evaluation stage. The three main experimental results of the MSTMM scheme are as follows: (1) the scheme of MSTMM can achieve template matching in only 0.015 seconds when a 64 × 64 template image slides on a 256 × 256 query image on a hardware platform with limited resources; (2) the matching success rate of the MSTMM scheme can reach up to 75% among 2107 experimental samples; and (3) the neural network training in the neural network mapping mechanism only takes at least 104.4 seconds on the CPU.INDEX TERMS Template matching, multimodal image, heterogeneous image, multisource image, visible and infrared image, image matching, neural network.

show abstract

“…In our experiments, we evaluate an algorithm using two criteria: matching Success Rate and algorithm runtime. We first define the matching accuracy, which is defined in (14).…”

Section: Experimental Results and Analysismentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Fast Template Matching Scheme of Visible and Infrared Image Under Occluded Scenarios

Mei

Wang²,

Wang

et al. 2022

IEEE Access

View full text Add to dashboard Cite

show abstract

“…In addition, there are differences in brightness according to the lighting of the two places where the camera is installed, or differences in brightness in an image according to the exposure of the aperture. In addition, various conditions cause problems that are difficult to correct [6]. To address these problems, many studies [7,8] have been conducted.…”

Section: Related Researchmentioning

confidence: 99%

Comparison of Single-Camera-Based Depth Estimation Technology for Digital Twin Model Synchronization of Underground Utility Tunnels

Park

Hong

Hwang³

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

Digital twin technology can be used for disaster safety management of underground utility tunnels, which are social infrastructures. To verify its efficiency in an underground utility tunnel, it must be implemented and monitored in the real world. Therefore, the state of information about an object and the position of an object in the real world, which can change upon movement, should be correspondingly reflected in the digital model. Underground utility tunnels are monitored through closed circuit television (CCTV) cameras because the main infrastructure facilities are installed in a long and narrow form, which limits access to them. Therefore, synchronization of a digital twin model of an underground utility tunnel requires a method for identifying the location of an object through an installed CCTV camera. This study compares the DenseDepth transfer learning method and the coordinate system conversion method through floor plan projection to estimate depth using a single CCTV camera installed in an underground utility tunnel. The coordinate system conversion method through floor plane projection showed an error of 0.23 m to 1.5 m within a 50 m range. The DenseDepth transfer learning method showed an error of 1 m to 3 m; therefore, it was considered unsuitable for distance estimation. The coordinate system conversion method through floor plane projection showed a smaller error even at a longer distance than the DenseDepth transfer learning method. Therefore, the coordinate system conversion method through floor plane projection was considered more suitable than the DenseDepth transfer learning method for depth estimation using a single camera.

show abstract