CN110991436A - Domestic sewage source separation device and method based on image recognition - Google Patents
Domestic sewage source separation device and method based on image recognition Download PDFInfo
- Publication number
- CN110991436A CN110991436A CN201911190906.3A CN201911190906A CN110991436A CN 110991436 A CN110991436 A CN 110991436A CN 201911190906 A CN201911190906 A CN 201911190906A CN 110991436 A CN110991436 A CN 110991436A
- Authority
- CN
- China
- Prior art keywords
- image
- module
- sewage
- dsp
- clear water
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/254—Fusion techniques of classification results, e.g. of results related to same input data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Image Processing (AREA)
Abstract
The method comprises the following steps that 1, a sewage sample and a clear water sample are used as training sets, and then the multi-features of the sewage sample and the clear water sample are used for classification; step 2, training by using an SVM classifier so as to obtain a classification template; step 3, classifying the sewage and clear water images to be classified through a template; step 4, correcting P classified by SVM by using moving object detection based on interframe difference method(Water)And P(wastewater)And performing fusion processing according to a D-S evidence theory synthesis rule to obtain a final decision fusion result. The invention can realize accurate identification of clear water and sewage by adopting an image processing mode, and quickly realize the separation of the clear water and the sewage by adopting hardware modules such as a DSP processor and the like, thereby ensuring the real-time property and the accuracy of the device.
Description
Technical Field
The invention relates to the technical field of environmental protection, in particular to a domestic sewage source separation device and method based on image recognition.
Background
At present, the common domestic sewage detection modes are mainly divided into three types. The first type: the discharge condition of the sewage port is determined mainly by monitoring 24-hour videos. However, due to the monitoring in this form, people inevitably generate mental stress due to long-term observation, so that the supervision work is not in place, and the water resource environment is polluted again. The second type: the judgment is carried out through the processes of extraction, assay, analysis and the like of the quality of the domestic sewage, but the quality of the domestic sewage is treated after being damaged in a large range, and the real-time property is obviously lacked. In the third category: the sensor is used for detecting the sewage flow, the processing program of the sensor changes along with the time change, the dynamic change of the sewage cannot be accurately mastered in real time, and the sensor is easy to corrode and damage after being placed in the sewage for a long time and has short service life. Therefore, a more intelligent means is promoted to be found for rapidly and accurately treating the domestic sewage.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention aims to provide the domestic sewage source separation device and method based on image recognition, which can realize accurate recognition of clear water and sewage by adopting an image processing mode, and quickly realize separation of the clear water and the sewage by adopting hardware modules such as a DSP (digital signal processor) and the like, thereby ensuring real-time performance and accuracy of the device.
In order to achieve the purpose, the invention adopts the technical scheme that:
domestic sewage source separator based on image recognition includes:
the video image acquisition module: the video image acquisition module is connected with the A/D conversion module and is used for collecting and acquiring a video image of sewage;
the light source lighting module: the light source illumination module is used for illuminating a sewage source head and mouth, overcoming the interference caused by the environment and providing powerful conditions for subsequent image processing;
an A/D conversion module: the A/D conversion module is respectively connected with the video image acquisition module and the video image compression processing module, and a video decoding chip is used for digitizing the collected image analog signals, namely the analog signals are converted into digital signals, so that a DSP (digital signal processor) is facilitated to carry out port processing;
the video image compression processing module: the video image compression processing module is respectively connected with the A/D conversion module and the DSP processor, and is used for compressing a large amount of digital video image data generated by A/D conversion and further providing a storage space for the DSP to process images;
a host interface module: the host interface module is connected with the DSP processor and is used for directly accessing the memory space of the CPU and accessing the storage of peripheral equipment;
a power supply module: the power supply module is respectively connected with the DSP processor and the video image acquisition and separation module and is used for providing +5V external input voltage to supply power to the DSP processor, the video image acquisition module and the separation module;
data memory card: the data storage card is connected with the DSP processor and is used for storing video image data and algorithm programs and storing black and white threshold values of preset values.
An image processing algorithm module: the image processing algorithm module is connected with the DSP, internally comprises an image enhancement module, an image graying module, an image binarization module and an image classification and identification module, and is used for realizing accurate identification of clear water and sewage images.
A reset circuit: the reset circuit is connected with the DSP and is used for initializing the state of the device into a null state;
a setting circuit: the setting circuit is connected with the DSP processor and is used for converting the logic value of the device into a specific value;
a bootstrap circuit: the bootstrap circuit is connected with the DSP processor and is used for superposing electronic elements such as a bootstrap boost diode, a bootstrap boost capacitor and the like with capacitor discharge and power supply voltage so as to increase the voltage;
a DSP processor: the DSP processor is positioned on the separating device and is respectively connected with the video image compression processing module, the host interface module, the power supply module, the data storage card, the image processing algorithm module, the reset circuit, the setting circuit and the bootstrap circuit; the system comprises a DSP (digital signal processor) module, an image processing algorithm module, a clear water image signal processing module and a sewage image signal processing module, wherein the DSP module is used for receiving data transmitted by the video image compression processing module, transmitting the data to the image processing algorithm module for processing, and defining 0 and 1 digital signals obtained by the DSP according to requirements, wherein 0 is a low level, namely the clear water image signal, and 1 is a high level, namely the sewage image signal;
a separation module: the separation module is positioned below the separation device and is respectively connected with the clean water pipeline and the sewage pipeline, and the clean water pipeline comprises: one end of the sewage discharge pipe is communicated with a sewage discharge source, and the other end of the sewage discharge pipe is communicated with a municipal sewage pipe for discharging cleaner sewage in domestic sewage; a sewage pipeline: set up one end and be linked together with the source of discharging sewage, the other end is linked together with the effluent water sump that sets up, and the effluent water sump is used for discharging excrement and urine.
The image processing algorithm module comprises a video image acquisition module, an image enhancement module, an image graying module and an image binarization module;
an image enhancement module: the image processing algorithm module is connected with the video image of the sewage and used for receiving the sewage image, the image shot by the camera is enhanced by an image processing method, the image characteristic of the sewage at the source port is emphasized, the color brightness of the sewage in the image is improved, the environmental characteristic around the pipeline port is inhibited, the difference between the environmental characteristic and the sewage is enlarged, the image quality is improved, and the identification effect is enhanced.
An image graying module: the image processing algorithm module is connected with the image enhancement module and used for receiving the enhanced sewage image and performing gray processing on the sewage image to obtain a gray sewage image;
an image binarization module: and the image processing algorithm module is connected with the image graying module, the gray value of each pixel of the gray image is respectively compared with a preset black-white threshold, when the gray value of the pixel is greater than the preset black-white threshold, the pixel is marked as a white pixel, and when the gray value of the pixel is less than the preset black-white threshold, the pixel is marked as a black pixel, so that the binary image is obtained.
The domestic sewage source separation method based on image recognition comprises the following steps;
an image classification and identification module: inside the image processing algorithm module, the image binarization module and the image acquisition module are connected, videos of sewage and clean water are converted into image sequence frames, 1000 samples of sewage and 1000 samples of clean water are obtained, wherein the clean water is selected as a positive sample, the sewage is selected as a negative sample, the samples are preprocessed and then classified by the SVM, and the steps of applying the SVM to classify are as follows:
step 1, using a sewage sample and a clear water sample as training sets, and then classifying by using multiple characteristics of the sewage sample and the clear water sample;
step 2, training by using an SVM classifier so as to obtain a classification template;
and 3, classifying the sewage and clear water images to be classified through the template.
Further applying moving object detection based on an interframe difference method, correcting P (clear water) and P (sewage) classified by the SVM, and carrying out fusion treatment according to a D-S evidence theory synthesis rule to obtain a final decision fusion result;
the method for counting the number of impurities in the image by moving object detection based on the interframe difference method is as follows: counting the number of all image impurities in the video sequence at intervals of set time T, recording the number as n, and correcting the classified result of the SVM by taking the n as a weight for distinguishing the clear water image and the sewage image;
step 1, taking a first frame image as an initial background;
step 2, graying the foreground image and the previous frame image, and carrying out difference subtraction to obtain a difference image;
step 3, setting a threshold value T as 15, and binarizing the difference image;
step 4, updating the background, namely updating the unchanged area to the background when the changed areas of the two previous frames of images are not updated, and then carrying out image difference between the foreground and the background again;
step 5, selecting the preset threshold value T as 15 again, and binarizing the difference image;
step 6, selecting 3 x 3 for binary differential image
The template of (2) is etched to eliminate the finely varying region.
Step 7, finding out an 8-connected region in the binarized differential image, and regarding the 8-connected region as a foreground motion region;
and 8, finding out the maximum area of the motion area, then extracting the impurity target in the area, and marking the impurity target in the foreground by using a rectangular frame.
Step 9 counts the number of impurities in the image.
The number of impurities in the image is detected by an interframe difference method, under an ideal condition, the number of the impurities in the clear water is small, and the number of the impurities in the sewage is large, so that the number of the impurities can be used as a weight, and the sewage and clear water image can be classified more accurately. And inputting the results of the SVM classifier and the interframe difference method moving object detection into DSP processing according to D-S decision fusion, wherein the low level (clear water image signal) is set as 0 by the DSP processing, and the high level (sewage image signal) is set as 1 by the DSP processing.
The separation module obtains a 0 signal, namely a clear water image, and a 1 signal, namely a sewage image, sent by the DSP, and the discharged clear water flows into a clear water pipeline, then is discharged into a municipal sewage pipeline, and the discharged sewage flows into a sewage pipeline, and finally is discharged into a self-made sewage pool.
The invention has the beneficial effects that:
the invention is used for separating clean water from sewage. Firstly, the device carries out video monitoring to the sewage source, then transmits the acquired image information to the image processing module for intelligent analysis and judgment, obtains the distinction of clear water and sewage, separates the judged result by a DPS processor, finally discharges the clear water to a domestic sewage pipeline, and discharges the sewage to a sewage pool. The device adopts the computer image processing technology, has realized the intelligent discernment of sewage state, has satisfied the requirement of sewage treatment real-time nature, can overcome traditional sewage device monitoring randomness poor, and intelligent degree is not high, and the construction cycle is long, defects such as with high costs. Meanwhile, as the sources of the clean water and the sewage are separated, the cleaner clean water can be continuously recycled, and the secondary pollution of the water is avoided. A large amount of sewage mixed with excrement can be used as a fertilizer for producing green food, thereby bringing great economic benefit to the society; according to the moving object detection decision fusion algorithm based on the SVM and the interframe difference method, firstly, the SVM performs primary judgment and identification on a sewage image, then the moving object detection based on the interframe difference method is re-identification on the sewage image and correction on an SVM output result, and finally, the moving object detection results of the SVM and the interframe difference method are fused according to a D-S evidence theory synthesis rule, so that a final decision fusion result is obtained, and the defect of low identification accuracy of a single classifier is overcome by adopting a mode of judging and identifying twice, so that the accuracy of sewage identification is improved.
Drawings
FIG. 1 is a schematic view of the overall structure of the domestic sewage source separation device based on image recognition.
Fig. 2 is a schematic diagram of a video image capture module.
FIG. 3 is a schematic diagram of an image processing algorithm module.
FIG. 4 is a schematic diagram of a moving object detection decision fusion algorithm based on SVM and interframe difference method.
Fig. 5 is a schematic diagram of the connection between the DSP processing module and the separation module.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings.
As shown in fig. 1, the domestic sewage source separation device and method based on image recognition comprises: the device comprises a video image acquisition module, a light source illumination module, a DSP processing module, an image processing algorithm module and a separation module. Wherein, DSP processing module includes: the video image compression device comprises an A/D conversion module, a DSP (digital signal processor), a video image compression processing module, a data storage card, a reset circuit, a setting circuit and a bootstrap circuit; the image processing algorithm module comprises: the system comprises an image enhancement module, an image graying module, an image binarization module and an image classification and identification module; the power supply module is used for providing +5V external input voltage to supply power to the DSP processor, the video image acquisition module and the separation module.
As shown in fig. 2, the light source illumination module is used to illuminate the source opening of the sewage, so as to overcome the interference caused by the environment, and ensure that the video image acquisition module acquires the best video image. The visible light camera in the video image acquisition module performs characteristic extraction on the state of the source mouth water of the sewage discharge pipeline, and then transmits the data of the visible light camera to the A/D converter module. And the analog signals collected by the visible light camera are converted into digital signals which can be identified by a DSP processor in the rear after being processed by the A/D conversion module. The converted digital signals are transmitted to a video image compression processing module, a large amount of data are compressed, and a storage space is provided for a DSP processor to process the data. And the compressed data is transmitted into the DSP processor. And finally, the DSP is responsible for forwarding the data to the image processing algorithm module connected with the DSP for judgment and identification.
As shown in fig. 3, the image processing algorithm module is connected to the DSP processor, and includes therein: the device comprises an image enhancement module, an image graying module, an image binarization module and an image classification and identification module. FIG. 4 shows a method for recognizing a sewage image based on SVM and interframe difference method. Firstly, training a classifier SVM, converting videos of sewage and clean water into image sequence frames, and obtaining 1000 samples of sewage and 1000 samples of clean water in total, wherein the clean water is selected as a positive sample, and the sewage is selected as a negative sample. The samples are preprocessed and then trained into a classifier SVM, then the trained classifier SVM is matched and classified with the acquired images, and the classification result is output in a probability form, namely P(Water)And P(wastewater)However, such classification is only a preliminary judgment of the clear water image and the sewage image, and if P exists(Water)0.51 and P(wastewater)In the case of such a blur of 0.49, it is obviously difficult to determine whether this is an image of clear water or sewage. Therefore, the moving object detection based on the interframe difference method is further applied to correct the P classified by the SVM(Water)And P(wastewater). Before moving object detection based on an interframe difference method is applied, images need to be preprocessed. First, the acquired image is acquiredThrough the processing of the image enhancement module, the color brightness of the discharged water in the image can be improved after the image enhancement, the environmental characteristics around the pipeline port are inhibited, the difference between the environmental characteristics is enlarged, the quality of the image is improved, and the recognition effect is enhanced. The image data is then transmitted to an image graying module. The image is converted into a gray image from an RGB color image, the gray image data is transmitted to binarization, and the image is changed into a binary image with pixel values of 0 and 255 from the gray image. And detecting the number of impurities in the image by using moving object detection based on an interframe difference method for the preprocessed image. Ideally, the amount of impurities in the clean water is small and the amount of impurities in the waste water is infinite. Thus, there is a critical value between the clear water image and the dirty water image. The critical value n is set to be 10, when the number n of impurities is larger than 10, the image can be judged to be a sewage image, and when the number n of impurities is smaller than 10, the image can be judged to be a clear water image. Because the classification result of the SVM is output in a probability form, and the impurity number n is a numerical value, the impurity number n → f (n) → δ → δ 'is converted, and finally the obtained δ' and the classification result of the SVM are fused by using a D-S evidence theory synthesis rule to obtain a final decision fusion result.
As shown in fig. 5, the DSP processing module is connected to the video image compression processing module, the image processing algorithm module, and the separation module. The DSP processor, an external data memory, a reset circuit, a bootstrap circuit and a setting circuit jointly form a DSP processing module, and the DSP processor is not only responsible for receiving data transmitted by the video image compression processing module and forwarding the data to the image processing algorithm module for judgment and identification, but also responsible for forwarding a result output by the image processing algorithm module to the separation module. And inputting the fusion result of the SVM classifier and the moving object detection decision of the interframe difference method into DSP processing in an image processing algorithm module, and outputting a result of 0, namely low level after the DSP processing if the probability that the fusion result belongs to the clear water image is high according to the setting. If the probability that the fusion result belongs to the sewage image is high, the output result is 1 after the DSP processing, namely, the high level. And then, the separation module receives the signals of 0 and 1 to carry out separation treatment, when the signal of 0 is received, the discharged water is judged to be clear water, the clear water flows into a clear water pipeline, and finally the clear water is directly discharged to a municipal sewage pipeline so as to be recycled and reused. When the signal 1 is received, the discharged water is judged to be sewage, the sewage flows into a sewage pipeline and is finally directly discharged into a sewage pool so as to develop human excrement into green fertilizer.
Claims (5)
1. Domestic sewage source separator based on image identification, its characterized in that includes:
the video image acquisition module: the video image acquisition module is connected with the A/D conversion module and is used for collecting and acquiring a video image of sewage;
the light source lighting module: the light source illumination module is used for illuminating a sewage source head and mouth, overcoming the interference caused by the environment and providing powerful conditions for subsequent image processing;
an A/D conversion module: the A/D conversion module is respectively connected with the video image acquisition module and the video image compression processing module, and a video decoding chip is used for digitizing the collected image analog signals, namely the analog signals are converted into digital signals, so that a DSP (digital signal processor) is facilitated to carry out port processing;
the video image compression processing module: the video image compression processing module is respectively connected with the A/D conversion module and the DSP processor, and is used for compressing a large amount of digital video image data generated by A/D conversion and further providing a storage space for the DSP to process images;
a host interface module: the host interface module is connected with the DSP processor and is used for directly accessing the memory space of the CPU and accessing the storage of peripheral equipment;
a power supply module: the power supply module is respectively connected with the DSP processor and the video image acquisition and separation module and is used for providing +5V external input voltage to supply power to the DSP processor, the video image acquisition module and the separation module;
data memory card: the data storage card is connected with the DSP processor and is used for storing video image data and algorithm programs and storing black and white threshold values of preset values;
an image processing algorithm module: and the image processing algorithm module is connected with the DSP and is used for realizing accurate identification of the clear water and sewage images.
A reset circuit: the reset circuit is connected with the DSP and is used for initializing the state of the device into a null state;
a setting circuit: the setting circuit is connected with the DSP processor and is used for converting the logic value of the device into a specific value;
a bootstrap circuit: the bootstrap circuit is connected with the DSP processor and is used for superposing electronic elements such as a bootstrap boost diode, a bootstrap boost capacitor and the like with capacitor discharge and power supply voltage so as to increase the voltage;
a DSP processor: the DSP processor is positioned on the separating device and is respectively connected with the video image compression processing module, the host interface module, the power supply module, the data storage card, the image processing algorithm module, the reset circuit, the setting circuit and the bootstrap circuit; the system comprises a DSP (digital signal processor) module, an image processing algorithm module, a clear water image signal processing module and a sewage image signal processing module, wherein the DSP module is used for receiving data transmitted by the video image compression processing module, transmitting the data to the image processing algorithm module for processing, and defining 0 and 1 digital signals obtained by the DSP according to requirements, wherein 0 is a low level, namely the clear water image signal, and 1 is a high level, namely the sewage image signal;
a separation module: the separation module is positioned below the separation device and is respectively connected with the clean water pipeline and the sewage pipeline, and the clean water pipeline comprises: one end of the sewage discharge pipe is communicated with a sewage discharge source, and the other end of the sewage discharge pipe is communicated with a municipal sewage pipe for discharging cleaner sewage in domestic sewage; a sewage pipeline: set up one end and be linked together with the source of discharging sewage, the other end is linked together with the effluent water sump that sets up, and the effluent water sump is used for discharging excrement and urine.
2. The domestic sewage source separation device based on image recognition as claimed in claim 1, wherein the image processing algorithm module comprises a video image acquisition module, an image enhancement module, an image graying module and an image binarization module;
an image enhancement module: the image processing algorithm module is connected with the video image of the sewage and used for receiving the sewage image, the image shot by the camera is enhanced by an image processing method, the image characteristic of the sewage at the source port is emphasized, the color brightness of the sewage in the image is improved, the environmental characteristic around the pipeline port is inhibited, the difference between the environmental characteristic and the sewage is enlarged, the image quality is improved, and the identification effect is enhanced.
An image graying module: the image processing algorithm module is connected with the image enhancement module and used for receiving the enhanced sewage image and performing gray processing on the sewage image to obtain a gray sewage image;
an image binarization module: and the image processing algorithm module is connected with the image graying module, the gray value of each pixel of the gray image is respectively compared with a preset black-white threshold, when the gray value of the pixel is greater than the preset black-white threshold, the pixel is marked as a white pixel, and when the gray value of the pixel is less than the preset black-white threshold, the pixel is marked as a black pixel, so that the binary image is obtained.
3. The domestic sewage source separation method based on image recognition is characterized by comprising the following steps;
an image classification and identification module: inside the image processing algorithm module, the image binarization module and the image acquisition module are connected, videos of sewage and clean water are converted into image sequence frames, 1000 samples of sewage and 1000 samples of clean water are obtained, wherein the clean water is selected as a positive sample, the sewage is selected as a negative sample, the samples are preprocessed and then classified by the SVM, and the steps of applying the SVM to classify are as follows:
step 1, using a sewage sample and a clear water sample as training sets, and then classifying by using multiple characteristics of the sewage sample and the clear water sample;
step 2, training by using an SVM classifier so as to obtain a classification template;
and 3, classifying the sewage and clear water images to be classified through the template.
4. The image recognition-based domestic sewage source separation method of claim 3, wherein moving object detection based on an interframe difference method is further applied, P (clear water) and P (sewage) classified by the SVM are corrected, fusion processing is performed according to a D-S evidence theory synthesis rule, and a final decision fusion result is obtained;
the method for counting the number of impurities in the image by moving object detection based on the interframe difference method is as follows: counting the number of all image impurities in the video sequence at intervals of set time T, recording the number as n, and correcting the classified result of the SVM by taking the n as a weight for distinguishing the clear water image and the sewage image;
step 1, taking a first frame image as an initial background;
step 2, graying the foreground image and the previous frame image, and carrying out difference subtraction to obtain a difference image;
step 3, setting a threshold value T as 15, and binarizing the difference image;
step 4, updating the background, namely updating the unchanged area to the background when the changed areas of the two previous frames of images are not updated, and then carrying out image difference between the foreground and the background again;
step 5, selecting the preset threshold value T as 15 again, and binarizing the difference image;
step 6, selecting binary differential image
The template of (2), the etching is carried out to eliminate the micro-variation area;
step 7, finding out an 8-connected region in the binarized differential image, and regarding the 8-connected region as a foreground motion region;
step 8, finding out the maximum area of the motion area, then extracting the impurity target in the area, and marking the impurity target in the foreground by using a rectangular frame;
step 9, counting the number of impurities in the image;
and inputting the results of the SVM classifier and the interframe difference method moving object detection into DSP processing according to D-S decision fusion, wherein the low level (clear water image signal) is set as 0 by the DSP processing, and the high level (sewage image signal) is set as 1 by the DSP processing.
5. The method as claimed in claim 4, wherein the separation module obtains a 0 signal (clear water image) and a 1 signal (sewage image) from the DSP, and the separation module discharges the clear water into the clear water pipeline and then into the municipal sewage pipeline, and discharges the sewage into the sewage pipeline and finally into the home-made sewage pool.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911190906.3A CN110991436B (en) | 2019-11-28 | 2019-11-28 | Domestic sewage source separation device and method based on image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911190906.3A CN110991436B (en) | 2019-11-28 | 2019-11-28 | Domestic sewage source separation device and method based on image recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110991436A true CN110991436A (en) | 2020-04-10 |
| CN110991436B CN110991436B (en) | 2023-04-07 |
Family
ID=70087825
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911190906.3A Active CN110991436B (en) | 2019-11-28 | 2019-11-28 | Domestic sewage source separation device and method based on image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110991436B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111495575A (en) * | 2020-04-26 | 2020-08-07 | 欧战备 | Swirler connecting pipe and blockage judging method |
| CN114764861A (en) * | 2022-04-19 | 2022-07-19 | 江苏禹润水务研究院有限公司 | Sewage treatment verification method based on computer vision |
| CN115856259A (en) * | 2022-12-12 | 2023-03-28 | 河南大学 | Water source conservation function remote sensing evaluation method and system based on regional difference characteristics |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107341790A (en) * | 2017-06-12 | 2017-11-10 | 广州大学 | A kind of image processing method of environment cleanliness detection |
| WO2018214974A1 (en) * | 2017-05-25 | 2018-11-29 | 中建钢构有限公司 | Method and system for displaying dynamic condition of pedestrians |
| CN109551662A (en) * | 2018-12-12 | 2019-04-02 | 谷学鑫 | A kind of household plastic waste recovery system |
-
2019
- 2019-11-28 CN CN201911190906.3A patent/CN110991436B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018214974A1 (en) * | 2017-05-25 | 2018-11-29 | 中建钢构有限公司 | Method and system for displaying dynamic condition of pedestrians |
| CN107341790A (en) * | 2017-06-12 | 2017-11-10 | 广州大学 | A kind of image processing method of environment cleanliness detection |
| CN109551662A (en) * | 2018-12-12 | 2019-04-02 | 谷学鑫 | A kind of household plastic waste recovery system |
Non-Patent Citations (1)
| Title |
|---|
| 陈喜胜等: "最大类间方差法在污水排放图像处理中的应用研究", 《现代计算机(专业版)》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111495575A (en) * | 2020-04-26 | 2020-08-07 | 欧战备 | Swirler connecting pipe and blockage judging method |
| CN114764861A (en) * | 2022-04-19 | 2022-07-19 | 江苏禹润水务研究院有限公司 | Sewage treatment verification method based on computer vision |
| CN114764861B (en) * | 2022-04-19 | 2022-10-28 | 江苏禹润水务研究院有限公司 | Sewage treatment verification method based on computer vision |
| CN115856259A (en) * | 2022-12-12 | 2023-03-28 | 河南大学 | Water source conservation function remote sensing evaluation method and system based on regional difference characteristics |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110991436B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110991436B (en) | Domestic sewage source separation device and method based on image recognition | |
| CN106251333B (en) | Element reverse detection method and system | |
| WO2018018788A1 (en) | Image recognition-based meter reading apparatus and method thereof | |
| US8050456B2 (en) | Vehicle and road sign recognition device | |
| CN106295789B (en) | Crop seed counting method based on image processing | |
| CN107038416B (en) | Pedestrian detection method based on binary image improved HOG characteristics | |
| CN109087363B (en) | HSV color space-based sewage discharge detection method | |
| CN104834912A (en) | Weather identification method and apparatus based on image information detection | |
| CN101916364A (en) | Adaptive dial digital identification device and method | |
| CN102305798A (en) | Method for detecting and classifying glass defects based on machine vision | |
| CN101788497B (en) | Embedded bean classifying system based on image recognition technology | |
| CN108009574B (en) | Track fastener detection method | |
| CN106960196B (en) | Industrial video small number recognition method based on template matching and SVM | |
| CN110335233B (en) | Highway guardrail plate defect detection system and method based on image processing technology | |
| CN101584624A (en) | Guideboard recognition blind-guide device and method thereof based on DSP | |
| CN106780428B (en) | Chip quantity detection method and system based on color recognition | |
| CN103177244B (en) | Method for quickly detecting target organisms in underwater microscopic images | |
| CN108563994B (en) | Parking lot parking space identification method based on image similarity | |
| CN115797351B (en) | Abnormality detection method for photovoltaic cell panel | |
| CN101008980A (en) | Method and system for automatic identifying butterfly | |
| CN107045633A (en) | A kind of digital locating segmentation method of the energy measurement table tool based on ST MSER | |
| CN116258999A (en) | Indicator light state detection method based on video stream | |
| CN114359864A (en) | Road surface disease detection method and system based on orthographic image and storable medium | |
| KR20100011187A (en) | Method of an image preprocessing for recognizing scene-text | |
| CN111488889B (en) | Intelligent image processor for extracting image edges |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |