A fast method for RGB to YCrCb conversion based on FPGA

Jiang, Hongxu; Li, Hanqing; Liu, Tingshan; Zhang, Ping; Lu, Jiazhen

doi:10.1109/iccsnt.2013.6967182

Cited by 7 publications

(5 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• Testing: Testing procedure is started with the trained weights and biases are loaded from the database and the hypothesis in calculated for each class using equation (3). The aim of this procedure is to obtain results of the classification performance.…”

Section: Feature Classification Phasementioning

confidence: 99%

“…The channels are based on luminance and two chromatic components. The three channels implementation is inspired by how human eyes are able to differentiate an object based on the shape, color, and brightness [3]. This proposed technique aims to extract richer and unique features of vehicles with the hypothesis that the unique features are able to differentiate the vehicle type based on the respective determined classes.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Vehicle Type Classification Using an Enhanced Sparse-Filtered Convolutional Neural Network With Layer-Skipping Strategy

2020

View full text Add to dashboard Cite

In this paper, a vehicle type classification approach is proposed by using an enhanced feature extraction technique based on Sparse-Filtered Convolutional Neural Network with Layer-Skipping strategy (SF-CNNLS). To extract rich and discriminant vehicle features, we introduce Three-Channels of SF-CNNLS (TC-SF-CNNLS) as the feature extraction technique. Local and global features of vehicles are extracted from three channels of an image which are, luminance and chromatic components. This technique is inspired by how human eyes differentiating objects that share almost similar features. TC-SF-CNNLS is tested with a benchmark dataset that provides frontal-view images to classify vehicle types of the bus, passenger car, taxi, minivan, SUV, and truck with Softmax Regression as a classifier. This test aims to observe the ability of this technique in differentiating vehicles with almost similar features but different classes. A test is also conducted with the self-obtained dataset (SPINT) to observe the effectiveness of this technique. The results are observed based on accuracy, precision, recall, and f-score, whereby, TCSF-NNLS has successfully recognized all the classes with an average accuracy of 0.905, precision is between 0.8629 to 0.9548, recall is between 0.83 to 0.96 and f-score is between 0.8564 to 0.9523. In addition, this technique is able to outperform other existing techniques with an average accuracy of 93.% compared to only 89.2% when 5 classes of vehicles are tested. INDEX TERMS Vehicle type recognition, convolutions neural network, deep learning, computational intelligence.

show abstract

Section: Feature Classification Phasementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Vehicle Type Classification Using an Enhanced Sparse-Filtered Convolutional Neural Network With Layer-Skipping Strategy

2020

View full text Add to dashboard Cite

show abstract

“…Researchers in [19,20,21] used the same approximation concept for converting color space in order to accelerate the conversion process. They converted the floating point numbers in the equations of converting color space into integers as a fast conversion method.…”

Section: Related Workmentioning

confidence: 99%

“…Chernov et al [19] used this approximation technique to convert color space from RGB to HSV (hue, saturation and value) and vice versa. They did not use lookup table technique unlike Jiang et al [20] where they converted the floating point numbers into integers and transformed multiplications to lookup tables and additions. They also used pipelining technology for farther acceleration.…”

Section: Related Workmentioning

confidence: 99%

“…In this paper, an approximation technique is applied to accelerate the conversion process. Based on approximation techniques used in [19,20,21], we found that it is possible to round the decimal points in (1), ( 2) and ( 3) into one decimal point digit. The new three equations for converting color space from RGB to YC b C r is shown as follows:…”

Section: A Fcca Stepmentioning

confidence: 99%

See 1 more Smart Citation

FPGA Implementation of Enhanced JPEG Algorithm for Colored Images

Abed¹

2020

IJCDS

View full text Add to dashboard Cite

Image quality and size have been growing very fast over the last decades requiring large storage space and high transmission rate. Image compression is an efficient method used to reduce the size of the image. JPEG algorithm has been considered as one of the famous techniques used for image compression. This paper proposed and implemented an optimized hardware solution called Hybrid Compression using Faster Color Conversion and Run Length (HC-FCC-RL) algorithm for JPEG algorithm based on FPGA to reduce the latency and accelerate the compression process. The paper also proposed a Fast Color Conversion with Approximation (FCCA) step to accelerate the conversion process from RGB to YC b C r . Using approximate techniques will reduce the number of resources used as well as the latency with some percentage error. In addition, the paper proposed a Parallel Run Length (P-RL) algorithm to speed up the design. The enhancement approach in this paper aimed to optimize the overall design of the JPEG algorithm targeting color images. To evaluate the performance of the proposed framework, the HC-FCC-RL architecture was implemented in Verilog and synthesized on FPGA board. The color conversion architecture uses 331 logic elements, a latency of 13.930 ns for converting the three colors component of blocks and a power dissipation of 861.03 mW. The percentage errors for the color conversion is 11.6%, 1.8% and 5.3% for Y, C b , C r components, respectively. The Run Length algorithm performs better than existing work by saving 48.36% in logic elements, 53.79% in latency and 62.86% in power dissipation. Thus, the proposed work demonstrated superior performance compared to current work in the literature.

show abstract