Novel L1 PCA informed K-means color quantization

Jaques, Lorenzo E.; Depoian, Arthur C.; Murrell, Ethan; Xie, Dong; Bailey, Colleen P.; Guturu, Parthasarathy

doi:10.1117/12.2619139

Cited by 3 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The results from the proposed nonlinear color assignment for each of the four methods is compared to linear color assignment 6 for 9 total bits on the Sailboat image in Figure 4. This equates to 3 bits per channel for the linear assignment and a varying number for the nonlinear assignment.…”

Section: Resultsmentioning

confidence: 99%

“…3 While initial work focused on this connection using L 2 -norm PCA, recently is was shown that using L 1 -norm PCA as calculated by the near optimal single bit flipping algorithm 4, 5 can improve results. 6 In both PCA K-means solutions (L 1 and L 2 ), color allocation has been assigned linearly, i.e. the same number of color bits for each R, G, and B channel.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

PCA informed non-linear channel color quantization

Depoian

Bailey

Murrell

et al. 2023

Multimodal Image Exploitation and Learning 2023

View full text Add to dashboard Cite

K-means is a popular unsupervised ML algorithm for analyzing and recognizing natural occurring patterns to cluster similar points together. When applied to the color space of an image, it can work to recognize segments of the image where more meaningful clustering can be applied. Color quantization has been employed for decades now to optimize the memory usage of saved images. Typical images are composed of red, blue, and green channels, each represented by a byte in memory. Therefore, each pixel can be represented by a total of 24 bits, resulting in around 16.8 million unique colors. However, the human perceptive system is not sensitive enough to require full usage of this color space and it is beneficial to find ways to reduce the number of colors closer to what the eye is able to distinguish. This results in more efficient use of memory while still preserving details and color separation in the image. The key issue is to determine how much a picture can be quantized before the image starts to degrade to the point that a human would be able to discern the difference. Currently there is no algorithm available to aptly determine where this point occurs or whether each color channel should be treated identically. This research applies K-means color clustering to each color channel of the image separately to optimize compression. The introduction of principal component analysis (PCA) informed K-means in place of randomly seeded K-means on each color channel separately further improves performance.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

PCA informed non-linear channel color quantization

Depoian

Bailey

Murrell

et al. 2023

Multimodal Image Exploitation and Learning 2023

View full text Add to dashboard Cite

show abstract

“…In classical dithering algorithms, calibration is usually achieved using perception-related functionals, such as the structural similarity index measure, that mimick human visual quality assessment [36,37]. Although very good results can be obtained with such metrics [38], the method lacks physical interpretability. Here, we wish to provide a sounder entropy-based approach that captures the ability of images to retain information during compression, as shown in recent visual appreciation experiments [27].…”

Section: Optimal Pixel Mappingmentioning

confidence: 99%

A new spin on color quantization

2023

View full text Add to dashboard Cite

We address the problem of image color quantization using a maximum entropy based approach. Focusing on pixel mapping we argue that adding thermal noise to the system yields better visual impressions than that obtained from a simple energy minimization. To quantify this observation, we introduce the coarse-grained quantization error, and seek the optimal temperature which minimizes this new observable. By comparing images with different structural properties, we show that the optimal temperature is a good proxy for complexity at different scales. Noting that the convoluted error is a key observable, we directly minimize it using a Monte Carlo algorithm to generate a new series of quantized images. Adopting an original approach based on the informativity of finite size samples, we are able to determine the optimal convolution parameter leading to the best visuals. Finally, we test the robustness of our method against changes in image type, color palette and convolution kernel.

show abstract

“…Another approach utilizes L1 principal component analysis (PCA)informed K-means to preserve the color definition of images [34]. Furthermore, a novel color quantization method based on an online k-means formulation has been developed, which utilizes adaptive and efficient cluster center initialization and quasi-random sampling to achieve high speed and high-quality quantization [35]. These methods demonstrate the effectiveness and efficiency of utilizing the K-means clustering algorithm for color quantization and feature extraction in computer vision and image processing.…”

Section: Employing Computer Vision and Image Processing For Aesthetic...mentioning

confidence: 99%

Enhancing the Harmonious Aesthetics of Architectural Façades: A VETAR Approach in Mengzhong Fort Village’s Stone Masonry

Chen,

Ji,

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

To enhance the continuity of character in the preservation of architectural heritage, this approach focuses on the horizontal self-similarity characteristics of architectural texture. A method using K-means and the Bhattacharyya approach for color selection in architectural repairs is proposed. It quantifies the visual coherence between the repaired structure and the original structure. Analyzing 12 images (A–L), with the original façade (image 0) as a reference, demonstrates that repairs using color-matched materials significantly improve visual coherence. Image A, created using the Visual Enhancement Through Adaptive Repair (VETAR) method, achieves the highest visual alignment with the original image. VETAR, grounded in Gestalt psychology, moves away from traditional materials to concentrate on visual consistency. Its successful implementation in the restoration of Mengzhong Fort illustrates a complex approach to material use in heritage conservation. After comparison, this method is deemed superior to traditional techniques, integrating modern interventions with historical aesthetics. The study suggests that VETAR may offer a referential method for architectural conservation, potentially facilitating a balanced integration of historical and contemporary elements in architectural renovation.

show abstract

Novel L1 PCA informed K-means color quantization

Cited by 3 publications

References 0 publications

PCA informed non-linear channel color quantization

PCA informed non-linear channel color quantization

A new spin on color quantization

Enhancing the Harmonious Aesthetics of Architectural Façades: A VETAR Approach in Mengzhong Fort Village’s Stone Masonry

Contact Info

Product

Resources

About