CN106646492B

CN106646492B - Fish finding method for analyzing sonar signals through image recognition

Info

Publication number: CN106646492B
Application number: CN201610855749.3A
Authority: CN
Inventors: 肖勇
Original assignee: Shenzhen Zhonglinghui Technology Co ltd
Current assignee: Shenzhen Zhonglinghui Technology Co ltd
Priority date: 2016-09-27
Filing date: 2016-09-27
Publication date: 2021-07-20
Anticipated expiration: 2036-09-27
Also published as: CN106646492A

Abstract

The invention discloses a fish detection method for analyzing sonar signals through image recognition. The invention has the advantages that: the underwater clutter interference is not easy to occur, the accuracy rate of detecting the fish school is high, and the production cost is reduced.

Description

Fish finding method for analyzing sonar signals through image recognition

Technical Field

The invention relates to the field of ultrasonic fish finders, in particular to a fish finding method for analyzing sonar signals through image recognition.

Background

The intelligent fish finder usually transmits sonar signals, reflects the signals when meeting objects moving underwater, and then judges the underwater fish condition by calculating the response time of the signals and the strength of the signals.

The general fish detection method is based on signal intensity and signal time series, detects whether fish appears and the size of the fish by judging signals reaching certain intensity in sonar signals and duration, and acquires the depth position of the fish in water according to the position of the signals appearing in the time series.

The analysis method based on the signal intensity is easily interfered by water bottom clutter and non-fish objects, and the judgment and detection of the fish are not accurate enough. Since the condition for determining the target fish is defined according to the threshold, the requirement for consistency of signal intensity is high. Based on the situation, most fish detecting equipment is added with a sensitivity adjusting function, the sensitivity needs to be adjusted manually to filter the background noise so as to adapt to different water quality environments, and the adjusted signal intensity can be ensured to adapt to the judgment condition of the target fish. This method therefore relies on the user's handling and, in the event of poor sensitivity adjustment, there will be excessive or undetectable fish.

Disclosure of Invention

The invention aims to provide a fish exploring method for analyzing sonar signals through image recognition, and the fish exploring method is used for solving the technical problem that the existing detection effect is not ideal.

The technical scheme of the invention is as follows: a fish finding method for analyzing sonar signals through image recognition is characterized by comprising the following steps: the method comprises the steps of synthesizing echoes collected by a sonar into continuous sonar signal images, and then identifying an arched image in the images to accurately detect the underwater fish school.

The bow-shaped image is an arc image which is smaller than a semicircle and is formed by a contract sonar signal image.

Further, the fish finding method comprises the following steps: step 1, a remote terminal receives sonar signals collected by a sonar probe; step 2, distributing corresponding color values according to the magnitude of the sonar signal intensity value; step 3, generating a row of longitudinal graph data in the memory every time one frame of data is acquired; step 4, splicing the acquired sonar image data according to the receiving time sequence; step 5, dividing an image rectangular area according to the size of a set window, and comparing and identifying the image rectangular area with a defined fish-bow-shaped image model; and 6, confirming that certain matching pairs are regarded as effective fish signals, and feeding the fish signals back to the display end to be displayed to the user.

Further, the work flow of the fish detecting equipment is as follows:

sonar fish finding equipment starts, step 1: emitting sonar excitation signals; step 2: collecting sonar echoes; repeating the step 1 and the step 2; and step 3: repeating the step 1 and the step 2 for a set number of times, and then wirelessly transmitting sonar data to the remote terminal equipment;

the remote terminal equipment is started, and step 4: receiving sonar data by remote terminal equipment; and 5: generating a frame signal image; step 6: synthesizing a frame signal image; and 7: dividing an identification area; and 8: comparing the synthesized frame signal image with the defined fish-bow-shaped image model; and step 9: successfully comparing the synthesized frame signal image with the defined fish-bow-shaped image model, and displaying a fish image; step 10: and (4) unsuccessfully comparing the synthesized frame signal image with the defined fish bow-shaped image model, returning to the step 4 to receive sonar data again, repeating the steps 4-10, and displaying a bow-shaped image capable of reflecting fish information on the remote terminal equipment after the set times are reached.

The fish swims in water, the feedback of the sonar to the target fish is a group of time-related signals, the signals are distinguished by colors according to the intensity through actual detection and analysis, and the signals are represented as images similar to a bow, so that the signal intensity in a certain area is not accurately judged by simply detecting the target fish.

The invention solves the defect that the traditional intelligent fish finder is inaccurate in fish finding due to the fact that signals are easily interfered by water quality. The traditional method needs manual adjustment of sensitivity, has poor operability, and is easy to cause the condition that fish cannot be detected or the fish is detected mistakenly once the adjustment is not good. The method provided by the invention corresponds the signals to the images, and even if the noise caused by the water quality is different, the trends of the signals are consistent, so that the shape of the target fish can be fed back on the images; compared with the traditional signal statistics method, the method based on image recognition is more in line with the dynamic model of the target fish and is not easily interfered by water quality

The invention provides a method for detecting fish signals based on image recognition, which synthesizes continuous sonar signal images and then achieves the effect of accurately detecting fish by recognizing an arch image in the images.

The analysis method based on the signal and image correspondence is not easy to be interfered by water bottom clutter and non-fish objects, and judges and detects fish accurately. Because the condition for judging the target fish is judged by the shape of the target fish fed back on the image, the requirement on the consistency of the signal intensity is low, therefore, the fish detecting method can filter bottom noise by omitting a sensitivity adjusting function, can adapt to different water quality environments, can ensure that different signal intensities are adapted to the judging condition of the target fish, does not depend on the processing of a user, and can accurately reflect the underwater fish swarm.

The invention has the advantages that: the underwater clutter interference is not easy to occur, the accuracy rate of detecting the fish school is high, and the production cost is reduced.

Drawings

FIG. 1 is a flow chart of fish signal identification according to the present invention;

FIG. 2 is a display interface of a fish school bow chart displayed by the fish finding method of the present invention;

in the figure: 1 is sonar fish-finding equipment; 2 is remote terminal equipment; 3. 4, 5, 6, 7, 8 and 9 are sonar signal image arcuate images of the fish.

Detailed Description

The technical scheme in the embodiment of the invention will be clearly and completely described below with reference to the accompanying drawings; it is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Example 1

A fish finding method through image recognition analysis sonar signal, first the echo that sonar gathers back synthesizes continuous sonar signal image, then through discerning the bow-shaped picture in the image to reach and detect the fish school under water accurately; the bow-shaped image is an arc image which is smaller than a semicircle and is formed by a contract sonar signal image.

Example 2

The fish finding method in the embodiment 1 can also be used for fish finding according to the following steps 1-6: step 1: the remote terminal receives sonar signals collected by the sonar probe; step 2: distributing corresponding color values according to the magnitude of the sonar signal intensity value; and step 3: generating a row of longitudinal graph data in the memory every time one frame of data is acquired; and 4, step 4: splicing the acquired sonar image data according to a receiving time sequence; and 5: dividing an image rectangular area according to the size of a set window, and comparing and identifying the image rectangular area with a defined fish-bow-shaped image model; step 6: and the fish signal which meets certain matching pair is regarded as an effective fish signal and is fed back to the display end to be displayed to the user.

Example 3

The fish detecting method as described in the above embodiment 1 or 2 may further include the following steps: sonar fish finding equipment starts, step 1: emitting sonar excitation signals; step 2: collecting sonar echoes; repeating the step 1 and the step 2; and step 3: repeating the step 1 and the step 2 for a set number of times, and then wirelessly transmitting sonar data to the remote terminal equipment;

the remote terminal equipment is started, and step 4: receiving sonar data by remote terminal equipment; and 5: generating a frame signal image; step 6: synthesizing a frame signal image; and 7: dividing an identification area; and 8: comparing the synthesized frame signal image with the defined fish-bow-shaped image model; and step 9: successfully comparing the synthesized frame signal image with the defined fish-bow-shaped image model, and displaying a fish image; step 10: and (4) unsuccessfully comparing the synthesized frame signal image with the defined fish bow-shaped image model, returning to the step 4 to receive sonar data again, repeating the steps 4-10, and displaying a bow-shaped image capable of reflecting fish information on the remote terminal equipment after the set times are reached. As shown in figure 1.

The fish bow map display interface displayed by the fish exploring method described in the above embodiments 1-3 is shown in fig. 2, in which

reference numerals

3, 4, 5, 6, 7, 8, and 9 are sonar signal image bow maps of fish.

The invention has the advantages that: the underwater clutter interference is not easy to occur, and the accuracy rate of detecting the fish school is high.

Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments or portions thereof without departing from the spirit and scope of the invention.

Claims

1. A fish finding method through image recognition analysis sonar signal, first the echo that sonar gathers back synthesizes continuous sonar signal image, then through discerning the bow-shaped picture in the image to reach and detect the fish school under water accurately;

step 1, a remote terminal receives sonar signals collected by a sonar probe; step 2, distributing corresponding color values according to the magnitude of the sonar signal intensity value; step 3, generating a row of longitudinal graph data in the memory every time one frame of data is acquired; step 4, splicing the acquired sonar image data according to the receiving time sequence; step 5, dividing an image rectangular area according to the size of a set window, and comparing and identifying the image rectangular area with a defined fish-bow-shaped image model; step 6, a certain matching pair is met, and the fish signal is determined to be an effective fish signal and fed back to a display end to be displayed to a user;

the working process of the fish finding equipment comprises the steps of starting the sonar fish finding equipment, and emitting sonar excitation signals; step 2, collecting sonar echoes; repeating the step 1 and the step 2; step 3, after repeating the step 1 and the step 2 for a set number of times, wirelessly transmitting sonar data to the remote terminal equipment;

starting up the remote terminal equipment, and step 4, receiving sonar data by the remote terminal equipment; step 5, generating a frame signal image; step 6, synthesizing a frame signal image; step 7, dividing the identification area; step 8, comparing the synthesized frame signal image with the defined fish-bow-shaped image model; step 9, successfully comparing the synthesized frame signal image with the defined fish-bow-shaped image model, and displaying a fish image; and 10, unsuccessfully comparing the synthesized frame signal image with the defined fish bow-shaped image model, returning to the step 4 to receive sonar data again, repeating the steps 4-10, and displaying a bow-shaped image capable of reflecting fish information on the remote terminal equipment after the set times are reached.