Compressed sampling nonlinear iterative optimization reconstruction method based on target plant
A technology of compressed sampling and iterative optimization, applied in image data processing, instruments, electrical components, etc., can solve the problems of long time consumption, large data, low reconstruction accuracy, etc., and achieve the effect of small step size
Pending Publication Date: 2019-06-07
JIANGSU UNIV
View PDF3 Cites 3 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
The purpose of the present invention is to propose a corresponding solution for the low reconstruction accuracy and long time consumption of the traditional compressed sensing method in agricultural image acquisition and reconstruction, so as to achieve effective observation and analysis of the growth status of some crops, Solve the storage and transmission problems caused by the large amount of crop observation data in the long-term phase
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0050] Such as figure 1 As shown, the overall flowchart of the nonlinear iterative optimization reconstruction algorithm based on machine vision compression sampling. The steps of specific implementation are as follows:
[0051] Step 1: For image acquisition, select the KinectV2 sensor, and the image acquired through the built-in SDK is an RGB image, and then convert it into the HSV color space to obtain the brightness map and tone map of the plant. details as follows:
[0052] Brightness map conversion formula: V=max(R, G, B) (1)
[0053] Tonemap conversion formula:
[0054]
[0055] Among them: R, G, and B are the red, green, and blue components in the RGB color space, respectively, and H ∈ [0, 360], R ∈ [0, 1], G ∈ [0, 1], B ∈ [0 , 1], V ∈ [0, 1].
[0056] Step 2: Use the Sobel edge detection algorithm to extract the overall shape and contour features of the target plant, and obtain the contour map of the plant.
[0057] Step 3: Acquisition of saliency feature ma...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention provides a compressed sampling nonlinear iterative optimization reconstruction method based on a target plant. The method specifically comprises the steps of: selecting a Kinect sensor to acquire a color image of a target plant in image acquisition; converting an RGB image into an HSV color space, obtaining a plant brightness feature map and a hue feature map, then obtaining a contour feature map of a plant by adopting a Sobel edge detection method, finally performing normalization processing on the feature map of each feature channel, and performing weighted average fusion to form a saliency feature map. In the area of econstruction algorithm, according to the method, a compressed sampling matching pursuit algorithm (CoSaMP) is used as a basis; regularization and variable step size adaptive ideas are fused; meanwhile, a Dog-Leg least square algorithm is combined to carry out iterative optimization. A regularization adaptive compression sampling matching pursuit algorithm(DLRaCSMP) based on Dog-Leg is provided, and the problems that in a traditional compression sampling algorithm, a support set is inaccurate, sparsity K is difficult to obtain, and precision is not enough and over estimation is caused by an SAMP fixed step length are solved.
Description
technical field [0001] The invention mainly relates to the fields of compressed sensing (Compressed Sensing, CS) and machine vision, and specifically relates to the field of agricultural plant image acquisition and compressed sensing reconstruction methods. Background technique [0002] Image acquisition and reconstruction is one of the key technologies in the development of machine vision technology. In recent years, with the continuous development of agricultural informatization and automation, image compression and reconstruction have played an important role in the detection of agricultural fruit plants, field management, collection of physiological parameters of plant microenvironment, and identification of greenhouse plant pests. At present, the demand for information continues to increase, and the signal bandwidth is getting wider and wider. How to compress, collect and reconstruct images with high speed and high quality has become a hot spot and focus of institutions...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): G06T7/12H04N19/42
CPCY02T10/40
Inventor 沈跃李尚龙刘慧黄忠裕吴边
Owner JIANGSU UNIV