CN113542762A

CN113542762A - Compression and enhancement method and system for vehicle infrared image

Info

Publication number: CN113542762A
Application number: CN202110822091.7A
Authority: CN
Inventors: 薛广月; 黄昌诚; 卢洋洋; 杨东凯; 李博闻
Original assignee: Zhongjiao Xinjie Technology Co ltd
Current assignee: Zhongjiao Xinjie Technology Co ltd
Priority date: 2021-07-20
Filing date: 2021-07-20
Publication date: 2021-10-22

Abstract

The invention discloses a method and a system for compressing and enhancing an infrared image of a vehicle, wherein the method comprises the steps of intercepting infrared image data from a video stream acquired by the vehicle, converting the hexadecimal image data into position information and pixel value size of each pixel point according to a transmission protocol, and forming an infrared bitmap image with a BMP format; compressing the infrared bitmap image to a size suitable for transmission to obtain a compressed image, and storing the compressed image in a memory for backup; the compressed image is converted to an 8bit bytecode according to base64 encoding. By utilizing the infrared imaging and image compression and enhancement principle, the problems of low precision, low speed and large image occupation space of a visual sensor in the Internet of vehicles are solved.

Description

Compression and enhancement method and system for vehicle infrared image

Technical Field

The invention relates to the technical field of image processing, in particular to a method and a system for compressing and enhancing an infrared image of a vehicle.

Background

The research on image compression coding dates back to the introduction of television signal digitization in 1984 for the first time. Until now, the image compression technique that has been used the longest and is also applied the most widely is the jpeg (joint Photographic Experts group) compression standard formally established in 1992, which is formally named "digital compression coding of information technology continuous tone still images". The method adopts a combined coding mode of predictive coding, discrete cosine transform and entropy coding, realizes higher compression ratio with lower algorithm complexity, and the compressed image also has good reconstruction quality. However, when the compression ratio is too high, the image quality is easily impaired, so JPEG is not suitable for displaying an image of high definition.

The traditional image compression coding takes information theory, digital signal processing and other theories as theoretical bases, research enters the mature period, objective and visual redundancy removing capability is close to the limit, and a bottleneck is difficult to break through by a plurality of main indexes such as coding efficiency, compression ratio, reconstruction quality and the like. With the continuous breakthrough of research of related emerging subjects, many new image compression algorithms are proposed, such as subband coding, fractal coding, model-based coding, wavelet transform image compression, image compression based on a neural network, and the like, which form a new generation of image compression coding technology.

The fractal concept is used for image compression and image description and processing based on fractal characteristics, the basic idea is that the characteristics of image internal self-similarity are utilized, an iteration function system is used for simulating and replacing an original image by using certain transformation, and when decompression is carried out, only a plurality of times of inverse transformation iteration are needed to be carried out on the image, so that a better and approximate reconstructed image can be obtained. Compared with JPEG standard compression, the fractal algorithm has better image compression quality, but the fractal algorithm has larger calculation amount and high time cost.

Wavelet Transform (Wavelet Transform) is used for the description of multi-resolution images, the basic idea of which is to decompose an image signal into a weighted sum of a cluster of translated and scaled basis functions. The wavelet transform is orthogonal transform without energy loss, can compress an image by adopting targeted non-uniform quantization, has a reversible algorithm, and can completely restore the image through reconstruction.

Image enhancement is a large important component of image processing, and the traditional image enhancement method plays an extremely important role in improving image quality. Since the last 90 s of the century, people have used digital techniques to improve images to formulate and analyze remote sensing images for efficient resource and mineral resource exploration, investigation, agricultural land use global planning and urban, crop yield estimation, weather forecasting, and disaster monitoring. In the field of biomedical engineering, image enhancement techniques are used to process radiological images, ultrasound images, and microscopic images of biological slices to improve the clarity and resolution of the images.

With the development of image technology research, new methods for image enhancement are continuously emerging. Some scholars have introduced fuzzy mapping theory in image enhancement algorithms to solve the problem of mapping function selection in enhancement algorithms. In addition, by applying the interactive image enhancement technology, the image enhancement effect can be subjectively monitored. Many new advances in image enhancement have been made using the stripe balance technique. For example, a multi-level bar-balancing algorithm, incorporating preservation of luminance, and a dynamic hierarchical bar-balancing algorithm are proposed. These algorithms can solve the problem of contrast range in bar graph equalization by segmenting the image and then balancing the sub-layer images.

In the prior art, a great deal of research proposals exist for image enhancement processing, but a great deal of problems to be solved in the aspects of identification, compression enhancement and the like of specific vehicle-mounted infrared images exist, and for example, a great deal of improvement space exists in the aspects of real-time performance of image processing, image precision, image compression ratio and the like.

Disclosure of Invention

The invention provides a compression and enhancement method and a compression and enhancement system for an infrared image of a vehicle, which can solve the problems of low precision and low speed of a visual sensor and large occupied space of the image in the Internet of vehicles.

According to one aspect of the invention, a method for compressing and enhancing an infrared image of a vehicle is provided, which comprises the following steps:

intercepting infrared image data from a video stream collected by a vehicle, converting the hexadecimal image data into position information and pixel value size of each pixel point according to a transmission protocol, and forming an infrared bitmap image with a BMP format;

compressing the infrared bitmap image to a size suitable for transmission to obtain a compressed image, and storing the compressed image in a memory for backup;

the compressed image is converted to an 8bit bytecode according to base64 encoding.

The compressing the infrared bitmap image to a size suitable for transmission to obtain a compressed image includes:

performing YCbCr color space transformation on the RGB color of the infrared bitmap image of the BMP to obtain a JPEG image;

performing DC level transfer on the JPEG image;

sub-sampling the YCbCr component of the JPEG image; the ratio of sampling YCbCr components is 4:1 or 4: 2;

carrying out N multiplied by N subblock blocking on the sub-sampled JPEG image according to pixel points;

performing discrete Fourier transform (DCT) on the NxN subblocks obtained by blocking the JPEG image, and converting image data into corresponding DCT coefficients;

performing quantization compression on the DCT coefficients;

performing Zig-zag scanning on the quantized and coded data;

coding an Alternating Current (AC) coefficient and a Direct Current (DC) coefficient of the data subjected to the zigzag-zigzag scanning;

and performing run length RLC coding on the AC coefficient.

And performing YCbCr color space transformation on the RGB color of the infrared bitmap image of the BMP according to the following formula:

Y＝0.299000R+0.587000G+0.114000B

Cb＝-0.169736R-0.331264G+0.500002B

Cr＝0.500000R-0.418688G-0.081312B

where Y is the luminance component, cb (u) and cr (v) are the blue-red chrominance components, and R, G and B are the colors of the red, green and blue channels, respectively.

The subblock partitioning of the sub-sampled JPEG image by NxN according to pixel points comprises the following steps:

and separating and storing 3 components which alternately appear and correspond to each pixel point in the JPEG image into 3 tables.

The discrete Fourier DCT is performed according to the following mode:

the forward transform formula of the one-dimensional DCT transform is:

where N, k is 0, …, N-1, N is the length of the image sequence before compression, and x (N) is the image sequence before compression;

the forward transform formula of the two-dimensional DCT is:

wherein, F (u, v) is the transformed image, F (x, y) is the original image, and C (u) and C (v) are the blue-red chrominance components;

the inverse transformation formula of the two-dimensional DCT is as follows:

the coefficients of the above formulae are: x, y, u, v ═ 0, 1

Where C (u) and C (v) are blue-red chrominance components.

The method further comprises the following steps:

intercepting infrared image data from a video stream sent to a data processor by an infrared camera through a vehicle acquisition module, converting hexadecimal image data into position information and pixel value size of each pixel point according to a transmission protocol, and forming an infrared bitmap image in a BMP format;

compressing the image to a size suitable for transmission by a compression module, and storing the image in a memory for backup;

the compressed image is converted into 8bit byte code by base64 coding, and the 8bit byte code is transmitted to the air by the transmitting station through an omnidirectional antenna and transmitted to the receiving station.

According to another aspect of the present invention, there is provided a compression and enhancement system for infrared images of a vehicle, the system comprising an infrared camera module, an acquisition module, a compression module, a memory, a transmitting station, and an omnidirectional antenna, wherein:

the infrared camera module is used for collecting video stream data;

the acquisition module is used for intercepting infrared image data from the video stream acquired by the infrared camera module, converting the hexadecimal image data into position information and pixel value of each pixel point according to a transmission protocol, and forming an infrared bitmap image in a BMP format;

the compression module is used for compressing the image to a compressed image with a size suitable for transmission, and storing the compressed image into the memory for backup;

the memory is used for saving and backing up the compressed image;

the transmitting radio station is used for converting the compressed image into 8bit byte codes according to base64 codes and sending the 8bit byte codes to the omnidirectional antenna;

the omnidirectional antenna is used for transmitting the compressed image.

The compression module specifically comprises:

the space change unit is used for carrying out YCbCr color space conversion on the RGB color of the infrared bitmap image of the BMP to obtain a JPEG image;

a DC shift unit for performing DC level shift on the JPEG image;

the sub-sampling unit is used for sub-sampling the YCbCr component of the JPEG image; the ratio of sampling YCbCr components is 4:1 or 4: 2;

the block dividing unit is used for carrying out N multiplied by N subblock division on the sub-sampled JPEG image according to pixel points;

the DCT transform unit is used for carrying out discrete Fourier DCT transform on the NxN subblocks obtained by blocking the JPEG image and converting the image data into corresponding DCT coefficients;

a quantization unit, configured to perform quantization compression on the DCT coefficient;

a Zig-zag scanning unit for Zig-zag scanning the quantized and encoded data;

the encoding unit is used for separately encoding an Alternating Current (AC) coefficient and a Direct Current (DC) coefficient for the data subjected to zigzag-zigzag scanning;

and the run-length code coding unit is used for performing run-length RLC coding on the AC coefficient.

And the space change unit performs YCbCr color space conversion on the RGB color of the infrared bitmap image of the BMP according to the following formula:

Y＝0.299000R+0.587000G+0.114000B

Cb＝-0.169736R-0.331264G+0.500002B

Cr＝0.500000R-0.418688G-0.081312B

Discrete Fourier DCT transformation in the DCT transformation unit is carried out according to the following mode:

the forward transform formula of the one-dimensional DCT transform is:

wherein N, k is 0.... times.n-1, N is the length of the image sequence before compression, and x (N) is the image sequence before compression;

the forward transform formula of the two-dimensional DCT is:

the inverse transformation formula of the two-dimensional DCT is as follows:

the coefficients of the above formulae are: x, y, u, v ═ 0, 1

Where C (u) and C (v) are blue-red chrominance components.

The technical scheme of the invention is adopted to provide a compression and enhancement scheme of the infrared image of the vehicle, the infrared image data is intercepted from the video stream collected by the vehicle, the hexadecimal image data is converted into the position information and the pixel value of each pixel point according to the transmission protocol, and the infrared bitmap image with the format of BMP is formed; compressing the infrared bitmap image to a size suitable for transmission to obtain a compressed image, and storing the compressed image in a memory for backup; the compressed image is converted to an 8bit bytecode according to base64 encoding. By utilizing the infrared imaging and image compression and enhancement principle, the problems of low precision, low speed and large image occupation space of a visual sensor in the Internet of vehicles are solved. The invention also has the advantages that: the compression rate is high, and the requirement on the communication channel bandwidth is low; the transmission rate is high, and the processing rate of a data processing center is improved; the hardware is simple to realize, the programmability is strong, and the subsequent upgrading and development are convenient.

The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:

FIG. 1 is a flow chart illustrating the principle of compressing and enhancing an infrared image of a vehicle according to an embodiment of the present invention;

FIG. 2 is a block diagram of an embodiment of the present invention illustrating a hardware design of an infrared image compression and enhancement system;

FIG. 3 is a flow chart of a JPEG encoder in accordance with an embodiment of the present invention;

FIG. 4 is an address sequence diagram of a Zig-zag scan according to an embodiment of the present invention;

FIG. 5 is a schematic diagram of run-length encoding according to an embodiment of the present invention;

fig. 6 is a schematic structural diagram of a compression and enhancement system for infrared images of a vehicle according to an embodiment of the present invention.

Detailed Description

The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.

Typically, a visible image is a planar energy distribution map, which may itself be a light-emitting object radiation source, or the energy reflected or transmitted by an object after it has been illuminated by a light radiation source.

Digital images can be represented by two-dimensional discrete functions:

I＝f(x，y)

where (x, y) represents the coordinates of the image pixel and the function value f (x, y) represents the grey value of the pixel at the coordinates.

It can also be represented by a two-dimensional matrix:

I＝A(M，N)

wherein, A is a matrix representation form, and M and N are matrix row and column lengths.

When an image is sampled, if each row of pixels is M and each column of pixels is N, the size of the image is M × N pixels, and thus A [ M, N ] forms an M × N real number matrix. The matrix element a (m, n) represents the pixel value of the image in the mth row and nth column, called a pixel or picture element.

The information for each pixel in a gray scale image is described by a quantized gray scale level, with no color information. The gray scale level of a gray scale image pixel is typically 8bits, i.e., 0-255. "0" indicates pure black, and "255" indicates pure white.

The image is numerically arbitrarily descriptive of pixel points, intensity and color. The storage capacity of the description information file is large, and the described object loses details or generates saw teeth in the scaling process. In the aspect of display, the color information of each point is presented in a digital mode after the object is resolved at a certain resolution, and the color information can be directly and rapidly displayed on a screen. Resolution and gray scale are the main parameters affecting the display.

Under the condition of meeting the requirement of certain fidelity, the image data is transformed, coded and compressed, redundant data is removed, and the data quantity required by representing the digital image is reduced, so that the image can be conveniently stored and transmitted. That is, a technique of expressing the original pixel matrix with a small amount of data with loss or without loss is also called image coding.

Image compression coding can be divided into two categories: one type of compression is reversible, i.e., the original image can be completely restored from the compressed data without loss of information, which is called lossless compression coding; another type of compression is irreversible, i.e. the original image cannot be completely restored from the compressed data, and there is a certain loss of information, which is called lossy compression coding.

In practical techniques, the total amount of image data can be compressed by:

(1) adopting a brightness (Y) and chroma (C) sampling mode;

(2) dividing the whole image into small areas for division processing;

(3) inter-frame and intra-frame data compression techniques are employed.

JPEG is the abbreviation of Joint Photographic Experts Group (Joint Photographic Experts Group), the post-file dropping name is ". jpg" or ". JPEG", is the most common image file format, is organized and established by a software development union, is a lossy compression format, can compress images in a small storage space, and repeated or unimportant data in the images can be lost, so the damage of image data is easily caused. Especially, the use of too high a compression ratio will significantly reduce the quality of the image restored after final decompression, and if a high quality image is sought, it is not suitable to use too high a compression ratio. However, the JPEG compression technique is very advanced, and it removes redundant image data by a lossy compression method, and can display very rich and vivid images while obtaining a very high compression rate, i.e., can obtain a good image quality with a minimum disk space. Moreover, JPEG is a very flexible format with the function of adjusting the image quality, allows files to be compressed with different compression ratios, supports multiple compression levels, the compression ratio is usually between 10: 1 and 40: 1, and the higher the compression ratio, the lower the quality; conversely, the smaller the compression ratio, the better the quality. For example, a 1.37Mb BMP bitmap file can be compressed to 20.3 KB. A balance point can also be found between image quality and file size. The JPEG format mainly compresses high-frequency information, well retains color information, is suitable for being applied to the Internet, can reduce the transmission time of images, can support 24-bit true colors, and is also generally applied to images needing continuous tone. After multiple comparisons, the 8 th level compression is adopted as the optimal proportion of both the storage space and the image quality.

Enhancing useful information in an image, which may be a process of distortion, is aimed at improving the visual impact of the image for a given image application. The method aims to emphasize the overall or local characteristics of the image, changes the original unclear image into clear or emphasizes certain interesting characteristics, enlarges the difference between different object characteristics in the image, inhibits the uninteresting characteristics, improves the image quality, enriches the information content, enhances the image interpretation and identification effects, and meets the requirements of certain special analysis.

The image enhancement method is to add some information or transform data to the original image by some means to selectively highlight interesting features in the image or suppress some unwanted features in the image to match the image with visual response characteristics. In the image enhancement process, the reason of image degradation is not analyzed, and the processed image is not necessarily close to the original image. The image enhancement technology can be divided into two categories, namely an algorithm based on a space domain and an algorithm based on a frequency domain according to different spaces of the enhancement processing process.

The spatial domain method is to operate the pixel points in the image, and is described by a formula as follows:

g(x，y)＝f(x，y)×h(x，y)

where f (x, y) is the original image; h (x, y) is a spatial transfer function; g (x, y) represents the processed image.

The algorithm based on the spatial domain directly operates the gray level of the image during processing, and the algorithm based on the frequency domain is an indirect enhancement algorithm which performs certain correction on the transformation coefficient value of the image in a certain transformation domain of the image. The algorithm based on the airspace is divided into a point operation algorithm and a neighborhood denoising algorithm. The point operation algorithm, namely gray level correction, gray level conversion, histogram correction and the like, aims to enable the image to be imaged uniformly, or expand the dynamic range of the image and expand the contrast. The neighborhood enhancement algorithm is divided into two types, namely image smoothing and sharpening.

Infrared images typically contain noise due to some kind of interference. The noise blurs the image and even submerges the characteristics, and if the noise is not processed in time, the subsequent processing process and even the output result are affected, and even an error conclusion can be obtained. Therefore, image noise filtering becomes an important component in the infrared image preprocessing. The smooth filtering of the space domain or the frequency domain can inhibit the image noise and improve the signal-to-noise ratio of the image.

When the neighborhood averaging method is adopted, the large template has large calculation amount, long time consumption, heavier fuzzy degree and poor image effect after processing. And with the increase of the template, the computation amount is rapidly increased, the time consumption is prolonged, the fuzzy degree is increased, and the image effect is poor. Therefore, it is preferable to use a smaller template when processing images in this way. When the median filtering method is adopted, the calculation amount of a large window is large, the time consumption is long, but the image effect after processing is basically equivalent to that of a small window. Moreover, as the window becomes larger, the larger the calculation amount, the longer the time consumption, but the improvement of the image effect is not obvious. Therefore, if the image is processed in this way, it is preferable to use a smaller window and use a faster algorithm. From the processed image, the image effect after filtering by the neighborhood averaging method is poor, the noise reduction is not obvious, and the image blurring degree is increased. The image after median filtering has good effect, clear image outline and greatly reduced noise, and the subsequent target identification, tracking and the like are more convenient. The median filtering not only better eliminates the effect of strong impulsive noise, but also better preserves the edges of the image. The gradient reciprocal smoothing algorithm has good filtering effect, but the operation time is too long, and the effect is good when the algorithm is applied to occasions with low real-time requirements. In view of the advantages of the median filtering, the median filtering is easy to implement in hardware, and can meet the requirement of real-time performance.

Fig. 1 is a flowchart illustrating compression and enhancement of an infrared image of a vehicle according to an embodiment of the present invention. As shown in fig. 1, the process of compressing and enhancing the infrared image of the vehicle includes the following steps:

step 101, intercepting infrared image data from a video stream acquired by a vehicle, converting the hexadecimal image data into position information of each pixel point and pixel value according to a transmission protocol, and forming an infrared bitmap image in a BMP format.

In the embodiment of the invention, the infrared image transmission system is based on the structure that the transmitting end consists of an infrared camera module, an acquisition module, a compression module, a memory, a transmitting radio station and an omnidirectional antenna; the receiving end consists of an omnidirectional antenna, a receiving radio station, a recovery module, an enhancement module, a memory and an upper computer display. Fig. 2 is a block diagram of a hardware structure of the transmitting end. Based on the system shown in fig. 2, the embodiment of the invention realizes compression and enhancement processing of the infrared image.

The invention designs a compression and enhancement system for an infrared image of a vehicle, which solves the problems of low precision, low speed and large image occupation space of a visual sensor in the Internet of vehicles by utilizing the principles of infrared imaging and image compression and enhancement. The embodiment of the invention mainly relates to an emission end of an infrared image transmission system. The transmitting terminal consists of an infrared camera module, an acquisition module, a compression module, a memory, a transmitting radio station and an omnidirectional antenna; the receiving end consists of an omnidirectional antenna, a receiving radio station, a recovery module, an enhancement module, a memory and an upper computer display.

The main work flow of the transmitting end of the compression and enhancement system for the infrared image of the vehicle designed by the invention is as follows: the system intercepts infrared image data from a video stream sent to a data processor by an acquisition module, converts the hexadecimal image data into position information and pixel value size of each pixel point according to a transmission protocol to form an infrared bitmap image in a BMP format, compresses the image to a size suitable for transmission by a compression module, stores the infrared bitmap image in a memory for backup, converts the compressed image into 8-bit byte codes by base64 coding, transmits the 8-bit byte codes to the air by a transmitting radio station through an omnidirectional antenna and transmits the 8-bit byte codes to a receiving radio station;

the infrared camera module uses an uncooled focal plane micro-thermal DM20 network temperature measurement type module which is divided into a front part and a rear part, wherein the front part is used for shooting and acquiring infrared images, a built-in chip is used for sorting and packaging the infrared images into a plurality of hexadecimal data packets, the data packets are output from an SPI _ MISO pin on the chip in the form of SPI signals and are sent to the rear part of the infrared detector; the second half is responsible for transmitting a plurality of received data packets from the RJ45 interface to the data processor through the network cable.

The acquisition module can use any upper computer capable of running a C + + program, such as a PC (personal computer), a raspberry pi (raspberry pi), a singlechip and the like. The invention uses a computer with a Windows10 system. The infrared detector is connected with the computer through a network cable, and the interface is a network port, namely RJ 45. The infrared detector and the computer are respectively provided with a power adapter for stably supplying power.

The image transmission module adopts a high-speed data transmission radio station of 19.2Kbps of a Nijing ND series, is provided with an MD192 type intelligent modem, adopts a digital signal processing DSP technology, realizes a wireless digital modulation and demodulation algorithm in real time in a software mode, and can set various parameters by software through AT instructions;

the compression module uses JPEG image compression algorithm, and the compression ratio is 10: 1.

The memory uses 16g flash memory chips.

And 102, compressing the infrared bitmap image to a size suitable for transmission to obtain a compressed image, and storing the compressed image in a memory for backup.

Fig. 3 is a flowchart of a JPEG encoder according to an embodiment of the present invention. The YCbCr color space adopted by JPEG supports 1-4 color components, and BMP is the RGB color space. The BMP image is compressed by first converting the color space. RGB, YUV, YCbCr, etc. have 3 color components; magenta, Cyan, Yellow and Black (Magenta, Cyan, Yellow, and Black) have 4 color components. The computer display image adopts an RGB color model, and the image data is formed by adding R, G, B three components. In this example, the YCbCr color space is used, and a color space conversion from RGB to YCbCr is performed. The conversion relationship between RGB and YCrCb is as follows:

Y＝0.299000R+0.587000G+0.114000B

Cb＝-0.169736R-0.331264G+0.500002B

Cr＝0.500000R-0.418688G-0.081312B

The image pixels are stored in unsigned integers. In an image, these sample data must be converted to two complementary codes to represent before any transformation or mathematical computation can be performed. DC level offset is to ensure that the sampling of the input data information is approximately centered around zero dynamic range. The method comprises the following steps: assuming that the sampling accuracy of the image component is n, each pixel value in the component should be subtracted by 2^n-1。

In general, the human eye is more sensitive to luminance variations than to color conversion, and thus, the Y component is more important than the Cb, Cr components. After the color space conversion, information of the image is mainly contained in the luminance component Y. A large amount of redundant color information is stored in the chrominance components Cb and Cr. Therefore, image compression can be achieved with a smaller amount of chrominance data being sub-sampled, losing a small amount of information. In the JPEG image standard, the sub-sampling format is typically 4:1 or 4: 2.

DCT (Discrete cosine Transform) is processed based on 8 × 8 sub-blocks, and therefore, the original image data needs to be divided into blocks before being transformed. Each pixel point in the original image is 3 components which appear alternately, and the 3 components need to be separated and stored in 3 tables.

DCT transform is a relatively common transform coding mode applied in code rate compression, and removes spatial redundant information through orthogonal transform to realize compression coding. The method comprises the steps of dividing source image data into N multiplied by N pixel blocks, and then utilizing DCT to carry out transformation operation on the pixel blocks one by one. In general, the energy of an image is concentrated in a low frequency region after discrete cosine transform, and thus DCT is mainly used to remove spatial redundancy of image data. After DCT, the correlation between coefficients of the image data may decrease, and most of the energy of the image data may be concentrated in a few DCT coefficients. DCT can effectively remove the correlation among image data and concentrate signal energy, is a core algorithm for realizing data compression by JPEG standard and is very important in image compression application. The DCT is evolved from a fourier transform.

The forward transform formula of the one-dimensional discrete cosine transform is as follows:

where N, k is 0.. times, N-1, N is the length of the pre-compression image sequence, and x (N) is the pre-compression image sequence.

The forward transform formula of the two-dimensional DCT is (N generally takes 8):

where F (u, v) is the transformed image, F (x, y) is the original image, and c (u) and c (v) are the blue-red chrominance components.

The inverse transformation formula of the two-dimensional DCT is as follows:

The coefficients of the above formulae are: x, y, u, v ═ 0, 1

Where C (u) and C (v) are blue-red chrominance components.

After the image data is converted into DCT coefficients, a quantization stage is performed to encode the image data. After the 8 × 8 image data is DCT-transformed, the high and low frequency energy is collected at the bottom right corner and the top left corner, respectively, as shown in tables 1 and 2.

TABLE 1 standard luminance quantization table

16	11	10	16	24	40	51	61
								12	12	14	19	26	58	60	55
14	13	16	24	40	57	69	56
								14	17	22	29	51	87	80	62
18	22	37	56	68	109	103	77
								24	35	55	64	81	104	113	92
49	64	78	87	103	121	120	101
								72	92	95	98	112	100	103	99

TABLE 2 standard chroma quantization Table

17	18	24	47	99	99	99	99
								18	21	26	66	99	99	99	99
24	26	56	99	99	99	99	99
								47	66	99	99	99	99	99	99
99	99	99	99	99	99	99	99
								99	99	99	99	99	99	99	99
99	99	99	99	99	99	99	99
								99	99	99	99	99	99	99	99

And (4) from the space domain, taking the upper left corner and removing the lower right corner, the realized function is similar to a low-pass filter. After quantization, the precision of the DCT coefficients is reduced, Alternating Current (AC) coefficients with relatively low effect in the image data are reduced, and the amount of image data is reduced (the AC coefficients represent image details), thereby achieving the purpose of image compression. Therefore, quantization is generally the most important step in image compression, and the main cause of image quality degradation is quantization. Quantization requires two frequency coefficients, luminance and chrominance are processed separately, and quantization is completed by dividing the quantization coefficient value obtained by the quantization matrix of each DCT coefficient and then rounding (usually rounding). The JPEG adopts uniform quantization, and the lower right corner has a larger value and the upper left corner has a smaller value as seen from a quantization table.

In general, the image compression effect achieved is different using different quantization tables. As shown in tables 1 and 2, a reference standard proposed by the JPEG standard, when used, can be adjusted by the user according to the characteristics of the source image and the performance of the image display. From the standard quantization table, it can be seen that the two tables are different for quantization and step size, thinner for luminance and thicker for chrominance. This is because the colors used for the image are in YUV format. The value of the chrominance (Y) component is relatively more important. We can separately perform a fine quantization on the luminance component (Y) and a coarse quantization on the chrominance component (UV), ensuring a higher compression ratio.

The JPEG standard specifies that 64 DCT coefficient values are stored at a time in the following order. This has the advantage that neighboring points in such a sequence are also adjacent in the image. The specific process is shown in fig. 4.

In the JPEG standard, the DC coefficient is encoded by differential Pulse modulation (dpcm) coding. The DC coefficient has two characteristics, one is that the coefficient value is higher; secondly, the coefficient values of two adjacent 8 × 8 image blocks are not changed greatly, and when a DC coefficient is encoded, the difference value of the DC coefficients of two adjacent blocks of the same image data is adopted: diff ═ DC_i-DC_i-1To difference valueThe encoding is implemented. This has the advantage that the difference is encoded in fewer bits than the original value.

The other values in the 8 x 8 image blocks are AC coefficients, using Run Length Coding (RLC). The quantized AC coefficients have a plurality of values of 0, and the data size can be effectively reduced by adopting run-length coding. The run-length encoded codeword is described by two bytes as shown in fig. 5.

Step 103, converting the compressed image into 8-bit byte code according to base64 coding.

In order to implement the above flow, the technical solution of the present invention further provides a compression and enhancement system for a vehicle infrared image, as shown in fig. 6, the compression and enhancement system for a vehicle infrared image includes an infrared camera module 21, an acquisition module 22, a compression module 23, a memory 24, a transmitting radio 25 and an omnidirectional antenna 26, wherein:

the infrared camera module 21 is used for collecting video stream data;

the acquisition module 22 is configured to intercept infrared image data from a video stream acquired by the infrared camera module, convert hexadecimal image data into position information of each pixel point and a pixel value according to a transmission protocol, and form an infrared bitmap image in a BMP format;

the compression module 23 is configured to compress the image into a compressed image with a size suitable for transmission, and store the compressed image in the memory for backup;

the memory 24 is used for saving and backing up the compressed image;

the transmitting radio station 25 is configured to convert the compressed image into an 8-bit bytecode according to base64 coding and send the 8-bit bytecode to the omnidirectional antenna;

the omnidirectional antenna 26 is used for transmitting the compressed image.

The compression module 23 specifically includes:

a DC shift unit for performing DC level shift on the JPEG image;

a Zig-zag scanning unit for Zig-zag scanning the quantized and encoded data;

Y＝0.299000R+0.587000G+0.114000B

Cb＝-0.169736R-0.331264G+0.500002B

Cr＝0.500000R-0.418688G-0.081312B

the forward transform formula of the one-dimensional DCT transform is:

the forward transform formula of the two-dimensional DCT is:

the inverse transformation formula of the two-dimensional DCT is as follows:

the coefficients of the above formulae are: x, y, u, v ═ 0, 1

Where C (u) and C (v) are blue-red chrominance components.

In summary, the technical solution of the present invention provides a compression and enhancement scheme for an infrared image of a vehicle, which captures infrared image data from a video stream collected by the vehicle, converts the hexadecimal image data into position information of each pixel point and pixel value according to a transmission protocol, and forms an infrared bitmap image in a BMP format; compressing the infrared bitmap image to a size suitable for transmission to obtain a compressed image, and storing the compressed image in a memory for backup; the compressed image is converted to an 8bit bytecode according to base64 encoding. By utilizing the infrared imaging and image compression and enhancement principle, the problems of low precision, low speed and large image occupation space of a visual sensor in the Internet of vehicles are solved. The invention also has the advantages that: the compression rate is high, and the requirement on the communication channel bandwidth is low; the transmission rate is high, and the processing rate of a data processing center is improved; the hardware is simple to realize, the programmability is strong, and the subsequent upgrading and development are convenient.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims

1. A compression and enhancement method for vehicle infrared images is characterized by comprising the following steps:

2. The method for compressing and enhancing the infrared image of the vehicle according to claim 1, wherein the compressing the infrared bitmap image into a size suitable for transmission to obtain a compressed image comprises:

performing DC level transfer on the JPEG image;

sub-sampling the YCbCr component of the JPEG image; sampling YCbCr component ratio of 4:1:1 or 4:2: 2;

performing quantization compression on the DCT coefficients;

performing Zig-zag scanning on the quantized and coded data;

and performing run length RLC coding on the AC coefficient.

3. The method as claimed in claim 2, wherein the YCbCr color space transformation is performed on RGB colors of the infrared bitmap image of the BMP according to the following formula:

Y＝0.299000R+0.587000G+0.114000B

Cb＝-0.169736R-0.331264G+0.500002B

Cr＝0.500000R-0.418688G-0.081312B

where Y is the luminance component, Cb and Cr are the blue-red chrominance components, and R, G and B are the colors of the red, green, and blue channels, respectively.

4. The method according to claim 2, wherein the sub-sampled JPEG image is sub-blocked by nxn sub-blocks according to pixel points, and the method comprises:

5. The method for compressing and enhancing the infrared image of the vehicle as claimed in claim 2, wherein the discrete fourier DCT transform is performed according to the following method:

the forward transform formula of the one-dimensional DCT transform is:

the forward transform formula of the two-dimensional DCT is:

the inverse transformation formula of the two-dimensional DCT is as follows:

the coefficients of the above formulae are: x, y, u, v ═ 0, 1

Where C (u) and C (v) are blue-red chrominance components.

6. The method of claim 1, further comprising:

7. A compression and enhancement system for infrared images of vehicles, the system comprising an infrared camera module, a collection module, a compression module, a memory, a transmitting radio and an omnidirectional antenna, wherein:

the infrared camera module is used for collecting video stream data;

the memory is used for saving and backing up the compressed image;

the omnidirectional antenna is used for transmitting the compressed image.

8. The system of claim 7, wherein the compression module specifically comprises:

a DC shift unit for performing DC level shift on the JPEG image;

a Zig-zag scanning unit for Zig-zag scanning the quantized and encoded data;

9. The system of claim 8, wherein the spatial transformation unit performs YCbCr color space transformation on RGB colors of the infrared bitmap image of BMP according to the following formula:

Y＝0.299000R+0.587000G+0.114000B

Cb＝-0.169736R-0.331264G+0.500002B

Cr＝0.500000R-0.418688G-0.081312B

10. The system of claim 8, wherein the discrete fourier DCT transform in the DCT transformation unit is performed according to the following:

the forward transform formula of the one-dimensional DCT transform is:

the forward transform formula of the two-dimensional DCT is:

the inverse transformation formula of the two-dimensional DCT is as follows:

the coefficients of the above formulae are: x, y, u, v ═ 0, 1

Where C (u) and C (v) are blue-red chrominance components.