CN105631874B - Utilize the method for image Segmentation Technology evaluation apple crisp slices browning degree - Google Patents

Abstract

The invention discloses a method for evaluating the browning degree of apple chips by image segmentation technology, which comprises: collecting the brightness value, red-green value and yellow-blue value of the outer surface of sample apple chips; Divide into 3 to 7 areas on average, each area is divided into 3 to 7 pigment blocks according to the range of red and green values, each pigment block is divided into 3 to 7 sub-pigment blocks according to the range of yellow and blue values, and the reference pigment block is screened out. The color range whose error between the definition and the RGB value of the reference pigment block is ≤±20% is the deep brown range; the present invention uses color electronic eyes to judge the degree of browning of apple crisps, which can not only avoid subjective evaluation, but also quickly obtain Color evaluation results, and it can also avoid the actual browning data of the product affected by the oxidation of the sample during pretreatment when chemical methods are used.

Description

Method for evaluating the browning degree of apple chips using image segmentation technology

technical field

The invention relates to the technical field of food detection, in particular to a method for evaluating the browning degree of crisp apple chips by using image segmentation technology.

Background technique

Apple crisps can not only maintain the original flavor and nutritional content, but also have the characteristics of crisp taste, green and natural, and easy storage, which meets consumers' needs for apple crisps, which are nutritious and convenient, naturally low in fat, and high in dietary fiber. Therefore, apple chips are very popular in developed countries such as Europe and America, and are mainly used as catering food, snack food, making fruit powder and instant drinks. However, in the production process of apple crisps, the browning problem restricts the further development of the apple crisps processing industry. The browning reaction not only restricts the grade of apple crisps products, but also seriously affects the sensory quality, flavor and nutrition of apple crisps products. Influenced, thereby restricting the appreciation of the apple crisp processing industry to a large extent, and bringing certain obstacles to the development of the apple crisp industry. Due to the difference in the degree of browning among different varieties during apple processing, it is very important to measure and compare the browning differences between varieties and under different technological conditions in apple variety breeding, resource evaluation and process selection.

At present, the evaluation methods for browning mainly include subjective grading method, spectrophotometer method and spectrophotometer. The subjective grading method is mainly to evaluate and classify the apparent color of the product through human eyes. This evaluation method does not rely on instruments, and places standard reference objects and products under a uniform standard light source. Sensory evaluators grade products according to different degrees of browning according to the reference objects. For example, apple chips are divided into grades 1, 2, 3, and 4 according to the degree of browning, with grade 1 being milder and grade 4 being severe browning. The spectrophotometer method is to measure the 420nm absorbance value (OD value) of the apple chips after being ground by a UV spectrophotometer, and evaluate the browning degree of the apple chips according to the obtained OD value. The larger the OD value, the browner the apple chips. The greater the degree of change. Compared with the traditional subjective evaluation method and spectrophotometer, the spectrophotometric color difference meter can obtain the L*a*b* value of apple chips through two detection modes of contact or non-contact for measurement, and realize rapid non-destructive testing of products.

There are some defects in the above three methods. In the subjective evaluation method, the quality requirements of the evaluators are high, and the evaluators need to be trained before they can participate in the evaluation, and because each person has different experience and experience, the evaluation results will also vary from person to person, and it is difficult to obtain a unified result, so there is a lack of Objectivity and repeatability. During the evaluation process of the spectrophotometric colorimeter, it is necessary to grind the processed apple chips. Even if the grinding is done in an ice bath, the sample will be in contact with the air, which cannot avoid oxidation and browning, resulting in inaccurate experimental results. The repeatability of the results was low, which affected the accurate assessment of the degree of browning of apple crisps. The spectrophotometric color difference meter can realize the non-destructive detection of the apparent color of apple chips, and quickly obtain the L*a*b* value and absorption spectrum, but the spectrophotometric color difference meter can only obtain the average value of the sampled local area of the material, and the shape of the material is also limited , it is difficult to fully reflect the apparent color distribution of the sample and the proportion of different browning degrees.

Contents of the invention

It is an object of the present invention to solve at least the above-mentioned problems and to provide at least the advantages which will be described later.

Another object of the present invention is to provide a method for evaluating the degree of browning of apple crisps using image segmentation technology. It uses a precision electronic eye as a tool to collect images of apple crisps and analyze the resulting images to evaluate the degree of browning of apple crisps. Overview.

In order to realize these purposes and other advantages according to the present invention, a kind of method utilizing image segmentation technology to evaluate the degree of browning of apple chips is provided, comprising:

Step 1, preparing a plurality of sample apple crisps, collecting the brightness value, red-green value and yellow-blue value of the outer surface of the sample apple crisps;

Step 2: Screen out the maximum and minimum values from the brightness values of all sample apple crisps, form a brightness range according to the maximum brightness value and the minimum brightness value of the sample apple chips, and divide the brightness range into 3-7 In each area, according to the range of red and green values in the area, the area is divided into 3 to 7 pigment blocks, and according to the range of yellow and blue values in each pigment block, the pigment block is evenly divided into 3 to 7 sub-pigment blocks, screen out the sub-pigment blocks with a cumulative contribution rate ≥ 85%, form a standard pigment block, screen out the standard pigment block with the largest RGB value, and use it as a reference pigment block to define the distance between the RGB value of the reference pigment block The color interval with error ±≤20% is the deep brown interval;

Step 3: collect the RGB values on the surface of the apple chips to be tested, obtain multiple RGB values to be tested, compare the RGB values to be tested with the RGB values in the deep brown range,

If the area corresponding to the RGB to be tested in the deep browning interval≤5% of the area of the apple crisp to be tested, it is judged that the apple crisp to be tested is mild browning;

If the area corresponding to the RGB to be measured in the deep browning interval accounts for 5% to 15% of the area of the apple crisps to be tested, it is judged that the apple crisps to be tested are moderately browned;

If the area corresponding to the RGB to be tested located in the deep browning interval is greater than or equal to 15% of the area of the apple crisps to be tested, it is determined that the apple crisps to be tested are severely browned.

Preferably, in the method for evaluating the degree of browning of apple crisps using image segmentation technology, in the first step, preparing multiple sample apple crisps, specifically: according to the shape of the apple raw material, the apples are cut into thicknesses It is sliced apple slices of 3-15mm.

Preferably, in the method for evaluating the degree of browning of crisp apple slices using image segmentation technology, in the first step, the apple slices are dried by hot air, vacuum drying, vacuum frying, microwave drying, vacuum freeze drying or variable temperature. Pressure difference puffing and drying to make the water content ≤ 7%, to obtain sample apple crisps.

Preferably, in the method for evaluating the browning degree of crisp apple slices using image segmentation technology, in the first step, the apple slices are dried at a temperature of 50-190°C until the water content is ≤7%, and the sample is obtained Apple crisp.

Preferably, in the method for evaluating the degree of browning of apple chips using image segmentation technology, in the first step, color electronic eyes are used to collect the brightness value, red-green value and yellow-blue value of the outer surface of the sample apple chips .

Preferably, in the method for evaluating the degree of browning of apple chips using image segmentation technology, in the second step, among the brightness values of all sample apple chips, the maximum value and the minimum value are screened out, according to the brightness value of the sample apple chips. The maximum luminance value and the minimum luminance value of the chips form a luminance range, and the luminance range is equally divided into 5 regions.

Preferably, in the method for evaluating the degree of browning of apple crisps using image segmentation technology, in the third step, the apple crisps to be tested are dried until their water content is ≤ 7%, and the surface of the apple crisps to be tested is collected. RGB value.

Preferably, in the method for evaluating the degree of browning of apple chips using image segmentation technology, in the third step, color electronic eyes are used to collect the RGB values of the surface of the apple chips to be tested.

Preferably, in the method for evaluating the degree of browning of apple chips using image segmentation technology, in the step 3, the RGB value to be measured is compared with the RGB value of the deep brown interval, specifically:

Calculate the error of all RGB values in the RGB to be measured and the depth brown interval, if there is an error between the RGB value and the RGB value to be measured in the depth brown interval ≤ ± 20%, then it is judged that the RGB value to be measured is located in the depth brown interval;

If there is no RGB value whose error with the RGB value to be measured is ≤±20% in the deep brown interval, it is determined that the RGB value to be measured is not located in the deep brown interval.

The present invention at least includes the following beneficial effects: In the present invention, the unified and objective reference standard for evaluating the browning degree of apple crisps established by using color electronic eyes makes the detection of apple crisps more accurate; the use of color electronic eyes can timely brown apple crisps Evaluation and grading of the degree of browning can avoid further errors in the process of sample preparation and measurement; the browning degree ratio of apple crisps can be obtained by using the color electronic eye, and the browning degree of dried apple crisps can be accurately quantified.

At the same time, the color electronic eye used in the present invention is available in the market, and the operation procedure is simple. The invention can provide a rapid method for evaluating the browning degree of apple chips for scientific research and food processing personnel, and can also provide reliable data support for browning grading.

Other advantages, objectives and features of the present invention will partly be embodied through the following descriptions, and partly will also be understood by those skilled in the art through research and practice of the present invention

Description of drawings

Fig. 1 is the flow chart of the present invention.

Fig. 2 is the image of the fresh sample apple slice that adopts common photographic equipment to collect in the embodiment 2 of the present invention.

Fig. 3 is the image of sample apple crisps collected by common photographic equipment in Example 2 of the present invention.

Fig. 4 is an image of fresh sample apple slices collected by color electronics in Example 2 of the present invention.

Fig. 5 is an image of a sample apple crisp collected by color electronics in Example 2 of the present invention.

Detailed ways

The present invention will be further described in detail below in conjunction with the accompanying drawings, so that those skilled in the art can implement it with reference to the description.

It should be understood that terms such as "having", "comprising" and "including" as used herein do not entail the presence or addition of one or more other elements or combinations thereof.

As shown in Figure 1, the present invention provides a kind of method utilizing image segmentation technology to evaluate the degree of browning of apple chips, comprising:

Step 1, preparing a plurality of sample apple crisps, collecting the brightness value, red-green value and yellow-blue value of the outer surface of the sample apple crisps;

Step 2: Screen out the maximum and minimum values from the brightness values of all sample apple crisps, form a brightness range according to the maximum brightness value and the minimum brightness value of the sample apple chips, and divide the brightness range into 3-7 In each area, according to the range of red and green values in the area, the area is divided into 3 to 7 pigment blocks, and according to the range of yellow and blue values in each pigment block, the pigment block is evenly divided into 3 to 7 sub-pigment blocks, screen out the sub-pigment blocks with a cumulative contribution rate ≥ 85%, form a standard pigment block, screen out the standard pigment block with the largest RGB value, and use it as a reference pigment block to define the distance between the RGB value of the reference pigment block The color interval with error ≤±20% is the deep brown interval;

Step 3: collect the RGB values on the surface of the apple chips to be tested, obtain multiple RGB values to be tested, compare the RGB values to be tested with the RGB values in the deep brown range,

If the area corresponding to the RGB to be tested in the deep browning interval≤5% of the area of the apple crisp to be tested, it is judged that the apple crisp to be tested is mild browning;

If the area corresponding to the RGB to be measured in the deep browning interval accounts for 5% to 15% of the area of the apple crisps to be tested, it is judged that the apple crisps to be tested are moderately browned;

If the area corresponding to the RGB to be tested located in the deep browning interval is greater than or equal to 15% of the area of the apple crisps to be tested, it is determined that the apple crisps to be tested are severely browned.

In one embodiment, in step 1, preparing multiple pieces of sample apple crisps specifically includes: cutting the apples into sliced apple slices with a thickness of 3-15 mm according to the shape of the raw material of the apples.

In one embodiment, in step 1, the apple slices are dried by hot air, vacuum drying, vacuum frying, microwave drying, vacuum freeze drying or variable temperature and pressure difference puffing to make the water content ≤ 7%, and the sample apples are obtained Crisp.

In one embodiment, in step 1, the apple slices are dried at a temperature of 50-190° C. to a water content of ≤7%, to obtain sample apple crisps.

In one embodiment, in step 1, color electronic eyes are used to collect the brightness value, red-green value and yellow-blue value of the outer surface of the sample apple crisp.

In one embodiment, in step 2, the maximum and minimum values are screened out from the brightness values of all sample apple crisps, and a brightness range is formed according to the maximum brightness value and the minimum brightness value of the sample apple crisps, Divide the brightness range equally into 5 zones.

In one embodiment, in step 3, the apple crisps to be tested are dried until their moisture content is ≤7%, and the RGB values of the surface of the apple crisps to be tested are collected.

In one embodiment, in step 3, color electronic eyes are used to collect the RGB values of the surface of the apple chips to be tested.

In one embodiment, in step 3, the RGB value to be measured is compared with the RGB value in the deep brown interval, specifically:

Calculate the error of all RGB values in the RGB to be measured and the depth brown interval, if there is an error between the RGB value and the RGB value to be measured in the depth brown interval ≤ ± 20%, then it is judged that the RGB value to be measured is located in the depth brown interval;

If there is no RGB value whose error with the RGB value to be measured is ≤±20% in the deep brown interval, it is determined that the RGB value to be measured is not located in the deep brown interval.

In order to make those skilled in the art understand the disclosed technology of the present invention more clearly, it will now be described in conjunction with embodiments.

Embodiment 1,

Electronic eye model: V270, manufacturer: American Hunterlab company, Digieye color electronic eye.

Six varieties of apples, namely Fuji, Qinglong, Yushang, Stark's Short Gold, Katina and Ruilin, were randomly selected as test materials, and 6 apples with uniform size and no pests and diseases were taken from each variety as raw materials for preparing sample apple chips. The sample apple crisps were prepared by flash drying with pulsating pressure difference.

Color electronic eye measurement: Turn on the light box, camera and computer, and perform standard whiteboard calibration. After the calibration is completed, perform color card calibration. After the calibration is completed, select the light source, select the diffuse reflection D65 light source, and warm up for 10 minutes. Place the sample in the working room. On the whiteboard, start the camera sample image acquisition, when the sample image is displayed, obtain the prepared sample apple crisp L* (brightness value), a* (red and green value) and b* through the image analysis and processing software CIE L*a*b* (yellow blue value) and other color evaluation indicators.

Image segmentation: According to the range of L* values of all sample apple chips, it is divided into 5 regions (45-54, 54-63, 63-72, 72-81, 81-90), each region, in Carry out 5 equal divisions according to the a* value to obtain 5 pigment blocks, and each pigment block is divided into 5 equal parts according to the b* value to obtain 5 sub-pigment blocks, and screen the sub-pigment blocks with a cumulative contribution rate ≥ 85%. In this embodiment There are five sub-pigment blocks whose cumulative contribution rate is ≥85%, and these five sub-pigment blocks are set as standard pigment blocks, and the standard pigment block with the largest RGB value is screened out as the reference color block, and the difference between the RGB value of the reference color block and the reference color block is defined. The color interval with an error of ≤±20% is the deep brown interval.

Then randomly reselect 6 varieties of apples, Fuji, Qinglong, Yushang, Stark Aujin, Katina and Ruilin as test materials to prepare sample apple crisps; put the prepared apple crisps to be tested in the color electronic In the eyes, the image is selected, the background is removed by using a filter, the RGB value of the apple chips to be tested is collected, and the browning ratio of the apple chips is obtained through image segmentation processing and image recognition technology, which can effectively make a comprehensive evaluation of the browning degree of the apple chips .

Table 1 shows the results of the electronic eye for the color of apple chips to be tested

It can be seen from Table 1 that the Catina apple has the largest color difference value △E, and its area ratio corresponding to the RGB value to be tested in the deep brown range accounts for 40.06% of the area of the entire apple crisp to be tested. Although the color difference value of Qinglong is not the largest, its deep brown ratio is the largest. Therefore, the color difference value cannot truly reflect the browning degree of apple crisps, and the deep brown area can better reflect the apparent crispness of apple chips. The ones with smaller color difference values are Fuji and Yugrain. The area proportions corresponding to the measured RGB values account for 4.73% and 10.57% of the area of the tested apple chips respectively, and the browning degree is lower than that of other varieties. Compared with sensory evaluation methods and spectrocolorimeters, the method provided by the invention is more accurate in evaluating the browning degree of the apparent color of apple crisps, with small errors, avoids subjective differences caused by human evaluation, and is convenient to operate and easy to grade.

Embodiment 2,

Randomly select Fuji apples as test materials, and take 6 apples with uniform size and no pests and diseases. In the experiment, different drying stages of the hot air-pulsating pressure difference flash were used to test the apple crisps.

Wash the apples, peel and core them, cut them into 5mm-thick discs, lay them flat on a tray, dry them with hot air at a temperature of 70°C until the moisture content of the apple slices is about 30%, and place them at 4°C 12 hours, placed in a flash tank, the flash temperature is 95°C, the number of flashes is 5 times, the evacuation temperature is 70°C, and the evacuation time is 1 hour. The sampling points during the drying stage of apple chips are fresh samples, and after the pre-drying is completed , after equalization, the pulsating pressure difference flash drying ends, and the sample apple crisps are obtained.

Determination with color electronic eyes: Turn on the light box, camera and computer. Carry out standard white board calibration. After the calibration is completed, use the color card for calibration. After the calibration is completed, select the light source, select the diffuse reflection D65 light source, and start the machine to warm up for 10 minutes. Place the sample on the working whiteboard and start the camera sample image acquisition. After the sample image is displayed, color evaluation indicators such as CIEL*a*b* are obtained through image analysis and processing software.

Image segmentation: According to the range of L* values of all sample apple chips, it is divided into 5 regions (45-54, 54-63, 63-72, 72-81, 81-90), each region, in Carry out 5 equal divisions according to the a* value to obtain 5 pigment blocks, and each pigment block is divided into 5 equal parts according to the b* value to obtain 5 sub-pigment blocks, and screen the sub-pigment blocks with a cumulative contribution rate ≥ 85%. In this embodiment There are five sub-pigment blocks whose cumulative contribution rate is ≥85%, and these five sub-pigment blocks are set as standard pigment blocks, and the standard pigment block with the largest RGB value is screened out as the reference color block, and the difference between the RGB value of the reference color block and the reference color block is defined. The color interval with an error of ≤±20% is the deep brown interval.

Then randomly re-select Fuji as the test material to prepare the sample apple crisps; place the prepared apple crisps in the color electronic eye, select the image, use the filter to remove the background, collect the RGB value of the apple crisps to be tested, and pass Image segmentation processing and image recognition technology can obtain the browning ratio of apple chips, which can effectively make a comprehensive evaluation of the browning degree of apple chips.

Table 2 shows the color electronic eye results of Fuji apple chips to be tested

Figure 2 and Figure 3 are pictures taken by ordinary photographic equipment. Figure 2 is the image of fresh sample apple slices, and Figure 3 is the image of sample apple crisps. The browning of the apple surface is most obvious during the steaming and drying stage, so the color difference value △E cannot well reflect the color change in the process of apple chips; Figure 4 is the fresh sample apple slices collected by the color electronic eye, and Figure 5 is the sample collected by the color electronic eye An image of an apple crisp, which can clearly capture the color difference of the sample apple crisp.

It can be seen from Table 2 that after analysis by color electronic eye image processing technology, it can be seen from the table that the proportion of the area corresponding to the RGB value to be tested in the deep brown range of the puffed apple crisps accounts for 4.73% of the area of the apple crisps to be tested , and the area ratio corresponding to the RGB value to be measured in the deep brown interval in the apple crisps to be tested after the wet treatment accounts for 0.29% of the area of the apple crisps to be tested. Compared with sensory evaluation methods and spectrocolorimeters, the method provided by the invention is more accurate in evaluating the browning degree of the apparent color of apple crisps, with small errors, avoids subjective differences caused by human evaluation, and is convenient to operate and easy to grade.

Although the embodiment of the present invention has been disclosed as above, it is not limited to the use listed in the specification and implementation, it can be applied to various fields suitable for the present invention, and it can be easily understood by those skilled in the art Therefore, the invention is not limited to the specific details and examples shown and described herein without departing from the general concept defined by the claims and their equivalents.

Claims (9)

1. A method utilizing image segmentation technology to evaluate the degree of browning of apple chips, characterized in that, comprising: Step 1, preparing a plurality of sample apple crisps, collecting the brightness value, red-green value and yellow-blue value of the outer surface of the sample apple crisps; Step 2: Screen out the maximum and minimum values from the brightness values of all sample apple crisps, form a brightness range according to the maximum brightness value and the minimum brightness value of the sample apple chips, and divide the brightness range into 3-7 In each area, according to the range of red and green values in the area, the area is divided into 3 to 7 pigment blocks, and according to the range of yellow and blue values in each pigment block, the pigment block is evenly divided into 3 to 7 sub-pigment blocks, screen out the sub-pigment blocks with a cumulative contribution rate ≥ 85%, form a standard pigment block, screen out the standard pigment block with the smallest RGB value, and use it as the reference pigment block, define the distance between the RGB value of the reference pigment block The color interval with error ≤±20% is the deep brown interval; Step 3: collect the RGB values on the surface of the apple chips to be tested, obtain multiple RGB values to be tested, compare the RGB values to be tested with the RGB values in the deep brown range, If the area corresponding to the RGB to be tested in the deep browning interval≤5% of the area of the apple crisp to be tested, it is judged that the apple crisp to be tested is mild browning; If the area corresponding to the RGB to be measured in the deep browning interval accounts for 5% to 15% of the area of the apple crisps to be tested, it is judged that the apple crisps to be tested are moderately browned; If the area corresponding to the RGB to be tested located in the deep browning interval is greater than or equal to 15% of the area of the apple crisps to be tested, it is determined that the apple crisps to be tested are severely browned. 2. the method for utilizing image segmentation technology to evaluate apple crisp browning degree as claimed in claim 1, is characterized in that, in described step 1, prepares many sample apple crisp slices, specifically: according to the shape of apple raw material, Cut the apples into slices with a thickness of 3-15 mm. 3. the method utilizing image segmentation technology to evaluate the degree of browning of apple crisps as claimed in claim 2, is characterized in that, in described step 1, apple slices is dried by hot air, vacuum drying, vacuum frying, microwave drying, Vacuum freeze-drying or variable-temperature-pressure-difference puffing-drying to make the water content ≤ 7% to obtain sample apple crisps. 4. The method for evaluating the browning degree of crisp apple slices using image segmentation technology as claimed in claim 2, characterized in that, in the first step, the apple slices are dried at a temperature of 50-190°C until the water content is ≤ 7%, get the sample Apple Crisp. 5. the method for utilizing image segmentation technology to evaluate the degree of browning of apple chips as claimed in claim 1 is characterized in that, in said step 1, the brightness value, red and green color of the outer surface of the sample apple chips are collected using color electronic eyes. value and yellow-blue value. 6. the method for utilizing image segmentation technology to evaluate the degree of browning of apple chips as claimed in claim 1, wherein in said step 2, in the brightness values of all sample apple chips, the maximum value and the minimum value are screened out. value, according to the maximum brightness value and the minimum brightness value of the sample apple chips, a brightness range is formed, and the brightness range is divided into 5 areas. 7. the method utilizing image segmentation technology to evaluate the degree of browning of apple chips as claimed in claim 1, characterized in that, in said step 3, the apple chips to be tested are dried to a water content≤7%, and collected to be tested The RGB value of the surface of the apple crisp. 8. the method for utilizing image segmentation technology to evaluate the degree of browning of apple chips as claimed in claim 7, characterized in that, in said step 3, the RGB value of the surface of the apple chips to be measured is collected with color electronic eyes. 9. the method utilizing image segmentation technology to evaluate the degree of browning of apple crisps as claimed in claim 1, it is characterized in that, in described step 3, the RGB value comparison of to-be-measured RGB value and deep brown interval is specially: Calculate the error of all RGB values in the RGB to be measured and the depth brown interval, if there is an error between the RGB value and the RGB value to be measured in the depth brown interval ≤ ± 20%, then it is judged that the RGB value to be measured is located in the depth brown interval; If there is no RGB value whose error with the RGB value to be measured is ≤±20% in the deep brown interval, it is determined that the RGB value to be measured is not located in the deep brown interval.