CN103235937A - Pulse-coupled neural network-based traffic sign identification method - Google Patents

Abstract

The invention discloses a pulse-coupled neural network-based traffic sign identification method. The method can be applied to the field of intelligent transportation, and is a means for assisting driving. According to the method, the color information of a traffic sign is fully utilized, a traffic sign image is converted into a hue, saturation and intensity (HIS) color space before being processed, a pulse-coupled neural network technology is used for extracting the multi-channel information of the traffic sign image on the basis of reserving the color information of the traffic sign to generate multi-channel characteristic sequences, the multi-channel characteristic sequences are combined into an integral characteristic sequence of the traffic sign, the integral characteristic sequence is used for matching in a traffic sign sequence template library, and a template with high overall relevance is determined as a correct identification result. Compared with the conventional pulse-coupled neural network technology-based grayscale image processing method, the method has the advantage that the important color information in the traffic sign is reserved, so that the identification rate is greatly increased.

Description

A kind of traffic sign recognition method based on Pulse Coupled Neural Network

Technical field

The present invention relates to a kind of traffic sign recognition method based on Pulse Coupled Neural Network, belong to intelligent transportation field, is a kind of means that can driver assistance.

Background technology

(1) Pulse Coupled Neural Network Feature Extraction Technology:

Artificial neural network is emerging subjects of nearly decades.It relates to multi-door subjects such as neuro-physiology, electronics, computer science, mathematics, has been widely used in numerous areas such as artificial intelligence, information processing, pattern-recognition, automatic control.Pulse Coupled Neural Network (Pulse-Coupled Neural Network, be called for short PCNN) be based on the neural network model that the research and development of the visual cortex neuron train of impulses synchronized oscillation phenomenon of cat is come, be called as third generation artificial neural network, compare with traditional artificial nerve network model, because its characteristic such as automatic propagation, synchronizing pulse granting with dynamic neuron, space-time summation characteristic, ripple receives much concern.In PCNN, neuron with similar input is provided pulse simultaneously, can remedy the subtle change on the discontinuous and amplitude in space of input data, thereby the area information of more complete reservation image, it has been used for by success that image smoothing, image are cut apart, aspects such as target is identified, feature extraction at present.This just makes PCNN have higher researching value and more wide application prospect.In recent years, the principle of work of PCNN is subjected to extensive attention with it both at home and abroad being applied in of field such as image processing, radar sonar, electron trade, medical and health, voice signal processing.

The PCNN model is to interconnect the feedback-type network that constitutes by several PCNN neurons, and each pixel of image can be regarded as a neuron, and each neuron is made up of three parts: importation, inner modulation device and pulse generator.

The neuronic triggering each time is referred to as igniting, and the PCNN model generates the feature duration of ignition of image by repeatedly iteration igniting.Each time in the iteration ignition process, the pixel of igniting can excite on every side the adjacent pixels point to light a fire, and outwards propagates thereby produce pulsating wave.By catching the image information of igniting pixel during iteration each time, can generate the sequence duration of ignition of image, utilizing the exclusive sequence signature duration of ignition of every width of cloth image to carry out image recognition is PCNN application process widely.

(2) RGB is to HSI color space switch technology:

Original colored traffic indication map similarly is based on the RGB color space, in image processing process, R, G, B component handled problems such as the color aberration that produces, distortion in order to reduce, this method proposes in preprocessing process, the RGB image transitions is handled to the HSI color space, made keeping finishing image processing process under the undistorted situation of color.

The HSI model is that U.S. chromatist Munsell (H.A.Munseu) proposed in 1915, and it has reflected the mode of people's vision system perception colour, comes aware colors with tone H, saturation degree S and three kinds of essential characteristic amounts of intensity I.The foundation of HSI model is based on two important facts: 1. the chromatic information of I component and image is irrelevant; 2. H and S component and people to experience the mode of color be allo.These characteristics make the HSI model be fit to very much color characteristic and detect and analyze.

(3) intelligent transport technology:

Intelligent transportation system (ITS) is unified traffic factors such as people, car, road to consider by appliance computer and infotech, is formed whole traffic administration and service.Intelligent traffic system is considered to be the effective way of social concerns such as the increase of transport solution accident quantity, road traffic congestion and environment, is expected to the highway communication cause of 21 century is produced positive impact.

Summary of the invention

Purpose of the present invention overcomes the deficiencies in the prior art exactly, and it is simple to provide a kind of hardware to realize, the traffic sign recognition method based on PCNN that discrimination is high.Traditional PCNN image-recognizing method is primarily aimed at gray level image and carries out the single channel feature extraction, neglected the colouring information of image, and the colouring information in the traffic sign is very important, the hyperchannel Pulse Coupled Neural Network traffic sign recognition method based on the HSI color space that the present invention proposes, can solve the problem that colouring information is lost in the image processing process, improved the discrimination of traffic sign greatly.

The technical solution used in the present invention is: a kind of traffic sign recognition method based on Pulse Coupled Neural Network, it is characterized in that, and may further comprise the steps:

Step 1: obtain the traffic sign original color image;

Step 2: described traffic sign original color image is transformed into the HSI color space from rgb color space;

Step 3: tone H, saturation degree S, the brightness I triple channel data of extracting described traffic sign original color image respectively;

Step 4: use the Pulse Coupled Neural Network technology that described tone H, saturation degree S, brightness I triple channel data are carried out the iteration ignition process, generate the triple channel pulse firing and compare sequence;

Step 5: with the triple channel pulse firing on described tone H, saturation degree S, the brightness I triple channel than combined sequence together, form the whole pulse firing of described traffic sign original color image than feature sequence;

Step 6: described whole pulse firing is carried out degree of correlation coupling than the characteristic sequence in sequence and the template base, identify correct road signs.

As preferably, described rgb color space is transformed into the HSI color space, and conversion formula is as follows:

$H=\left\{\begin{array}{c}\frac{\mathrm{\π}}{3}\×\frac{G-B}{\mathrm{Max}},\mathrm{ifMax}=R\\ \frac{\mathrm{\π}}{3}\×\frac{B-R}{\mathrm{Max}-\mathrm{Min}}+\frac{2\mathrm{\π}}{3},\mathrm{ifMax}=G\\ \frac{\mathrm{\π}}{3}\×\frac{R-G}{\mathrm{Max}-\mathrm{Min}}+\frac{4\mathrm{\π}}{3},\mathrm{ifMax}=B\end{array}\right.$

H=H 10 π, ifH＜0

$S=\left\{\begin{array}{c}\frac{\mathrm{Max}-\mathrm{Min}}{\mathrm{Max}+\mathrm{Min}}.0<I\&le;\frac{1}{2}\\ \frac{\mathrm{Max}-\mathrm{Min}}{2-(\mathrm{Max}+\mathrm{Min})},I>\frac{1}{2}\end{array}\right.$

$I=\frac{1}{2}(\mathrm{Max}+\mathrm{Min})$

Wherein, H, S, I are the values of tone, saturation degree and three components of brightness of HSI color space; R, G, B are the values of three components of rgb color space red, green, blue, and wherein the value of R, G, B is normalized between 0～1; (B) (B), namely Max and Min get maximal value and minimum value to .Min=min to Max=max for R, c for R, c.

As preferably, described PCNN model is to interconnect the feedback-type network that constitutes by several PCNN neurons, and each neuron is made up of importation, inner modulation device and pulse generator.

As preferably, described neuron is each time in the iteration ignition process, the pixel of igniting can excite on every side the adjacent pixels point to light a fire, thereby producing pulsating wave outwards propagates, by catching the information of described traffic sign original color image iteration time igniting each time pixel, can generate described triple channel pulse firing and compare sequence.

Compare with traditional neural network image recognition methods, based on the image-recognizing method of PCNN have efficiently, fast, high, the hardware of discrimination realizes characteristics such as simple.

With existing image-recognizing method contrast based on PCNN, the hyperchannel Pulse Coupled Neural Network traffic sign recognition method of the HSI color space that the present invention proposes has the following advantages:

The main traffic sign of China has red, yellow, blue, black, white five kinds of colors at present.Traditional PCNN traffic sign recognition method is converted into gray level image to the traffic sign image and carries out feature extraction, has lost the important colouring information of traffic sign, is easy to cause the mistake identification of traffic sign.Hyperchannel Pulse Coupled Neural Network traffic sign recognition method based on the HSI color space, carry out the PCNN image characteristics extraction on the basis that keeps the traffic sign colouring information, improve the discrimination of traffic sign greatly, played crucial effects for correct driver assistance.

Description of drawings

Fig. 1: be the techniqueflow of the hyperchannel Pulse Coupled Neural Network traffic sign recognition method that the present invention is based on the HSI color space.

Fig. 2: be rgb color space model of the present invention and HSI color space model synoptic diagram.

Fig. 3: be Pulse Coupled Neural Network neuron models synoptic diagram of the present invention.

Embodiment

Below in conjunction with the drawings and specific embodiments a kind of traffic sign recognition method based on the hyperchannel Pulse Coupled Neural Network that the present invention proposes is described further.

Ask for an interview Fig. 1, the technical solution adopted in the present invention is: a kind of traffic sign recognition method based on Pulse Coupled Neural Network may further comprise the steps:

Step 1: obtain the traffic sign original color image.

Step 2: described traffic sign original color image is transformed into the HSI color space from rgb color space;

Ask for an interview Fig. 2, be rgb color space model of the present invention and HSI color space model synoptic diagram; Rgb color space is to the conversion based on the polar bicone of cylinder by a unit cube based on Descartes's rectangular coordinate system to the conversion of HSI color space, basic demand is that the luminance factor in the rgb color space is separated, colourity is decomposed into the color harmony saturation degree, and show tone with the angle scale, conversion formula is as follows:

$H=\left\{\begin{array}{c}\frac{\mathrm{\&pi;}}{3}\&times;\frac{G-B}{\mathrm{Max}},\mathrm{ifMax}=R\\ \frac{\mathrm{\&pi;}}{3}\&times;\frac{B-R}{\mathrm{Max}-\mathrm{Min}}+\frac{2\mathrm{\&pi;}}{3},\mathrm{ifMax}=G\\ \frac{\mathrm{\&pi;}}{3}\&times;\frac{R-G}{\mathrm{Max}-\mathrm{Min}}+\frac{4\mathrm{\&pi;}}{3},\mathrm{ifMax}=B\end{array}\right.$

H=H 10 π, ifH＜0

$S=\left\{\begin{array}{c}\frac{\mathrm{Max}-\mathrm{Min}}{\mathrm{Max}+\mathrm{Min}}.0<I\&le;\frac{1}{2}\\ \frac{\mathrm{Max}-\mathrm{Min}}{2-(\mathrm{Max}+\mathrm{Min})},I>\frac{1}{2}\end{array}\right.$

$I=\frac{1}{2}(\mathrm{Max}+\mathrm{Min})$

Wherein, H, S, I are the values of tone, saturation degree and three components of brightness of HSI color space; R, G, B are the values of three components of rgb color space red, green, blue, and wherein the value of R, G, B is normalized between 0～1; (B) (R, c are that Max and Min get maximal value and minimum value B) to .Min=min to Max=max for R, c.

Step 3: tone H, saturation degree S, the brightness I triple channel data of extracting described traffic sign original color image respectively.

Step 4: use the Pulse Coupled Neural Network technology that described tone H, saturation degree S, brightness I triple channel data are carried out the iteration ignition process, generate the triple channel pulse firing and compare sequence;

Ask for an interview Fig. 3, be Pulse Coupled Neural Network neuron models synoptic diagram of the present invention; Described PCNN model is to interconnect the feedback-type network that constitutes by several PCNN neurons, and each neuron is made up of importation, inner modulation device and pulse generator.Each pixel of described traffic sign original color image can be regarded as a neuron, and (i, j) individual neuron is that example illustrates neuronic interaction in the Pulse Coupled Neural Network to present embodiment with.I _IjRepresentative (i, j) neuronic outside stimulus input, Y _IjRepresent neuron (i, output j), U _IjRepresent neuron (i, internal activity item j).The importation is made up of the two large divisions, is respectively feedback channel input F _IjWith linearity link input L _IjT _IjBe dynamic threshold, β is the strength of joint coefficient between the synapse.Pulse produces part and is made up of threshold adjuster, comparer, pulse producer. as internal activity item U _IjGreater than dynamic threshold T _IjThe time, the PCNN neuron produces output Y _IjWhen neuron has pulse output, excite dynamic threshold T _IjSharply increase, the increase of thresholding has guaranteed that this neuron can not produce the output of pulse for the second time at once, does not produce pulse output and causes thresholding to begin to be lower than internal activity item U according to the index law decay when threshold value drops to again _IjThe time, begin to have pulse output again, and then threshold value go round and begin again carry out above-mentioned variation.The output of pulse affects other neuronic output as other neuronic input again.As output valve Y _Ij(n gets 1, claims the neuron igniting; Work as Y _Ij(n) get 0, claim that neuron misfires; Described neuron is each time in the iteration ignition process, the pixel of igniting can excite on every side the adjacent pixels point to light a fire, thereby producing pulsating wave outwards propagates, by catching the information of described traffic sign original color image iteration time igniting each time pixel, can generate described triple channel pulse firing and compare sequence; This method is extracted in each iterative process, the ratio of the pixel of igniting and total pixel, i.e. and iteration igniting is than the characteristic sequence of sequence as traffic sign.

Step 5: with the triple channel pulse firing on described tone H, saturation degree S, the brightness I triple channel than combined sequence together, form the whole pulse firing of described traffic sign original color image than feature sequence;

Step 6: described whole pulse firing is carried out degree of correlation coupling than the characteristic sequence in sequence and the template base, identify correct road signs.

Specific embodiment described herein only is that the present invention's spirit is illustrated.Those skilled in the art can make various modifications or replenish or adopt similar mode to substitute described concrete enforcement, but can't depart from spirit of the present invention or surmount the defined scope of appended claims.

Claims (4)

1. the traffic sign recognition method based on Pulse Coupled Neural Network is characterized in that, may further comprise the steps:

Step 1: obtain the traffic sign original color image;

Step 2: described traffic sign original color image is transformed into the HSI color space from rgb color space;

Step 3: tone H, saturation degree S, the brightness I triple channel data of extracting described traffic sign original color image respectively;

Step 4: use the Pulse Coupled Neural Network technology that described tone H, saturation degree S, brightness I triple channel data are carried out the iteration ignition process, generate the triple channel pulse firing and compare sequence;

Step 5: with the triple channel pulse firing on described tone H, saturation degree S, the brightness I triple channel than combined sequence together,

Form the whole pulse firing of described traffic sign original color image than feature sequence;

Step 6: described whole pulse firing is carried out degree of correlation coupling than the characteristic sequence in sequence and the template base, identify correct road signs.

2. the traffic sign recognition method based on Pulse Coupled Neural Network according to claim 1, it is characterized in that: described rgb color space is transformed into the HSI color space, and conversion formula is as follows:

$H=\left\{\begin{array}{c}\frac{\mathrm{\&pi;}}{3}\&times;\frac{G-B}{\mathrm{Max}},\mathrm{ifMax}=R\\ \frac{\mathrm{\&pi;}}{3}\&times;\frac{B-R}{\mathrm{Max}-\mathrm{Min}}+\frac{2\mathrm{\&pi;}}{3},\mathrm{ifMax}=G\\ \frac{\mathrm{\&pi;}}{3}\&times;\frac{R-G}{\mathrm{Max}-\mathrm{Min}}+\frac{4\mathrm{\&pi;}}{3},\mathrm{ifMax}=B\end{array}\right.$

H=H 10 π, ifH＜0

$S=\left\{\begin{array}{c}\frac{\mathrm{Max}-\mathrm{Min}}{\mathrm{Max}+\mathrm{Min}}.0<I\&le;\frac{1}{2}\\ \frac{\mathrm{Max}-\mathrm{Min}}{2-(\mathrm{Max}+\mathrm{Min})},I>\frac{1}{2}\end{array}\right.$

$I=\frac{1}{2}(\mathrm{Max}+\mathrm{Min})$

Wherein, H, S, I are the values of tone, saturation degree and three components of brightness of HSI color space; R, G, B are the values of three components of rgb color space red, green, blue, and wherein the value of R, G, B is normalized between 0～1; Max=max (R, G, BD, (B), namely Max and Min get maximal value and minimum value to Min=min for R, G.

3. the traffic sign recognition method based on Pulse Coupled Neural Network according to claim 1, it is characterized in that: described Pulse Coupled Neural Network model is to interconnect the feedback-type network that constitutes by several Pulse Coupled Neural Network neurons, and each neuron is made up of importation, inner modulation device and pulse generator.

4. the traffic sign recognition method based on Pulse Coupled Neural Network according to claim 3, it is characterized in that: described neuron is each time in the iteration ignition process, the pixel of igniting can excite on every side the adjacent pixels point to light a fire, thereby producing pulsating wave outwards propagates, by catching the information of described traffic sign original color image iteration time igniting each time pixel, can generate described triple channel pulse firing and compare sequence.

Images