Topic: Side Imaging Understanding Distances and more......

[RGecy]

Ok, here is a diagram by Tom Vickers from Humminbird (with a few enhancements from me) that I think will help aid in and maybe stir a little discussion.  There are some very interesting objects shown and how they will appear on the Side Imaging view using 455khz. 

I think whats also interesting is that this diagram was drawn long before Down Imaging was even thought of!  Notice how the object directly under the boat is displayed as beeing cut in half.

Robert

[sonar2000]

Good picture Robert.  Unless Hummimbird will tell us what they mean by 86 degrees coverage and explain the 4 degree piece we are having to find out ourselves.  On another note. with DI it may not be a big deal any longer.  We can see side to side and also directly down.  If we use the 2 screen display (SI & DI) we have all the bases covered. I have looked at some past recordings and on the Viewer (HV) the directly under the boat is most times a split view (left & right) and shows a piece of a bottom tsrget and sometime it is on both SI screens. If the 2d down is also displayed the GPS location matches so between the two screens we will see directly under the boat.  but we need spring to get here so we can get on the lake and really run this with DI and recording.  HB has a great product.  We have to learn how to use it.  Chuck

[mrj9600]

Hi Robert
Great info.
Is this part of some official information from Humminbird ? And if yes, where can I find the whole document ?

I think I will make a few adjustments in the HumViewer program cause with this new information I think I can make it a little more accurate.

/Martin

[RGecy]

Martin this is the whole document.  I added a few more fish to help understand how fish in the water column can look like they are in the same position but actually are farther apart.

What do you think you would be able to make more accurate from this?  Waypoints?

Robert

[Stymie]

This is good information.  Totally opened my eyes to this technology...but does this mean I need to take my hs trig book out on the waters??

[sonar2000]

I think you will be ok with out it.  lol   the area of question is really small and  with the DI views we are looking at the bottom and using the two screen display we are looking at the undern the boat bottom with DI.  Math makes my small brain hurt so I wont wish that on anyone .   however I do have a small excel program calculator that figures the (possible) missed coverage under the boat on SI.   all you have to do is enter your depth.  It will give you the cone at that depth.  Heck if it indeed exists we can always add a behind the boat outrigger to compliment our side outriggers.  chuck

[Stymie]

A fella on another forum has pulled a similar document out but is discounting the DI showing what's under the boat part. 

Still interesting.  It seems that every day, my assumptions of this technology are challenged and I learn new things.

[RGecy]

Stymie,

They can challenge all they want!  As they say, the proof is in the pudding!  I have documented several instances personally where the SI coverage at 455khz shows directly beneath the boat, and long before Down Imaging was even around! 

Robert

[Kimi]

Im no engineer, but there is something that is not matching here! 
If you look at my picture bellow, you can see the waves on the surface. The cursor mark is on the wreck, that is laying 61, 7 m from the keel-line to the left. And I know for sure that the rocks on the right síde is about 140-200m south of the wreck. If the thinny white line betwean the "blind area" and the visible area should be the keel-line, all the mesurements would not be relevant. The wreck was marked with two bouys at the moment, one on each end. And I navigated useing to GPS systems, one of them was a chartplotter. I also got a papermap of this old divesite, so I know how it should look like.
 
If you would cut away the "blind/black area", and that is what you should do, if you seriously means that there are no "blind sector" in this picture.

In that case the distans betwean the wreck and the rocks should be only about 30-40m. That should be great! Because then we could anchor our boats on the underwater rocks, where we could get a wery good grip for the anchor, and we could dive down to wreck from there. But the distance is more than that, so we have to dropp the anchor on the wreck.
 
Still very sceptic...Kimi   

[sonar2000]

I have just been corrected on the DI.       This is a display using the 455khz SI freq and not the 2d 200 khz as earlier stated .  maybe my one cell brain is over worked.  ok. back to the school books and what ever else I can find.  Chuck

[RGecy]

Kimi,

Measuring horizonal distance between objects in an SI image is a little tricky, especially when you get deeper like this.The area closer to the center is compressed because the slant ranges are very close to each other in length. Even though the distance only appears to be a short ways apart, you have to recitify the image to get an accurate representation "Visually"!  You cant just remove the water column and the image is visually correct.  What you are seeing here is called slant range distortion.

Now, this may take more than I can fathom to explain, so I am going to copy some articles off the net that explain this in detail.

Slant Range Distortion   http://www.l-3klein.com/?page_id=77
The ranges measured by a basic side scan sonar record are slant ranges from the sonar to each point recorded (see “Sonar Geometry” section). Unless an automatic correction is applied to the data, these will also be the ranges when the data is displayed. Inherent in a slant range display are two principal distortion factors which should always be kept in mind and will often need to be corrected by some means.

The first is, the distances to features on the record are not horizontal distances to either side of the sonar. That is, we cannot look at a record and directly measure the offset of a feature from the ship’s track. This must be found by calculation, and will be treated separately under the heading “Horizontal Offset Calculation”.

Secondly, targets are compressed in the across-track direction. The difference in slant ranges to the leading and trailing edges of a target will always be less than the actual extent of the target. The degree of compression varies with the horizontal distance to the target. The closer the target to the sonar, the more severe will be the compression. At greater distances the slant ranges are closer to the true horizontal ranges and the compression effect is less severe. Actual target extent in the across-track direction can be found by the same calculations used for true horizontal offsets. The effect is shown schematically in the illustration below.



The records below show a portion of sonar image before and after correction for slant range and speed distortions. Note that the feature closer to the sonar is changed more by the correction process than the features further away.

Details

Horizontal Offset Calculation
Many side scan sonar applications require that we be able to accurately pinpoint the locations of targets seen on the records. A basic side scan sonar system (that is, one not including an automatic correction opt ion) displays slant ranges from the sonar to features on the bottom. For accurate positioning we must determine the horizontal distance from the sonar to these targets. As discussed in the “Slant Range Distortion” section, the distorting effect is greatest at the shorter ranges from the sonar.

Horizontal offset can be found by calculation using the Pythagorean theorem. Referring to the above schematic, In the following example, the towfish height is one leg of the right triangle, measured as 15 meters. The measured slant range is the hypotenuse,measured as 27 meters, and the horizontal offset or range is the other leg (side scan sonar always assumes a flat bottom). The horizontal range is then:

Details

I hope this helps,

Robert

[sonar2000]

wow Robert,  about 5 years from now we will look back on this and  say how very simple.  and then we can get a job at NASA as a rocket scientist.  I think some of us push the envelope a bit more than others but this is where we learn.  and thanks to this forum no one gets upset about thing they dont understand and keep moving forward by asking the next question.  Many thanks to those who ask and to those who answer.  I can hardly wait to get the break in the weather and get on the water.  Side imageing is amazing and for what we can now see is fantastic.  Chuck

[Stymie]

I guess I'm confused again.

Doesn't HB have the ability to create a waypoint from an object found on SI.

How does it know where that object is?  How does it know where in the water column that object is?  Or is it assuming that the object is on the bottom?

[RGecy]

Yes, the unit does the calculations for you!  It does assume that the object you are marking is on the bottom!  So if you mark a fish that is in the water column, the waypoint may not be accurate!

Robert

[Stymie]

IF I were to mark something of interest on sidescan, it would most likely be structure so that I could go over later to investigate.

I also see me marking weed lines and then coming back to fish up and down the weed line.

So with that, I don't see this as an issue.

If I see some fish on sidescan while trolling, now I've got a problem, cuz now I know they are there but not sure where. So if they don't bite, is it my presentation or me fishing in the wrong spot or both!

[RGecy]

Stymie,

Like you said, most of the time you are marking the structure that the fish are on.  But if you do mark something that is in the water column it may be off a little.  How much depends on where the object is.  Objects closer to the surface will be less accurate than objects near the bottom.  The further they are out the horizontal range and the slant range become closer and therfore more accurate.

Robert

[RGecy]

This is a post I made on another forum where the SI coverage and DI is being highly debated.  I wanted to certainly share my thoughts here and get some feedback from our members as well.  It is quiet long, and I didn't expect to write a book, but here goes! 

------------------------------------------------------------------------------------------------------------------------

Ok, I have poured over this conversation for the last hour or so and thought I might throw my hat in the ring for a few rounds.  I am a loyal HB owner and supporter since the beginning.  I was one of the first to decode the HB format and have written software to view the HB data on the PC as well as decoded the Lowrance format and know the workings intimately.  I have personally built Side Scan Sonars and potted my own transducers, and studied Side Scan Sonars years before HB came out with Side Imaging.  I do not consider myself an expert, and in now way claim to be.  But lets just say, I am not a novice!

The problem here is that most people do not understand exactly how a Side Scan Sonar element even works. Or how the data is stored and processed in the HB or Lowrance units.  I will try to shed some light on my thoughts on the subject!

The first debate is weather or not the Humminbird SI covers 180 degrees. I can definitely say, at 455kHz, it does cover 180 degrees, especially in shallow water less than 75'.  I and others here have proven without doubt that HB SI will see the same object on both sides using 455kHz.  The proof is in the pictures!  I have been making this argument for years, long before DI and SS even came about!  Rickard offered some scientific evaluations to this on another thread. 

A side imaging beam for one does not have the razor straight edges that are drawn in diagrams by Humminbird or even the nicely drawn diagrams in this post.  Those are drawn for the layman or novice so it is easy to understand how it works in theory.  It is not a scientific representation by any means.  A Side scan sonar beam is actually shaped more in a rounded fan shape in the vertical and has a very narrow beam in the horizontal.  Even in the horizontal it is not this razor beam.  It actually has some angle to it, be it very small, and also has what are called side lobes.  Even though the HB literature states 86 degrees in the vertical beam, the sound still projects past this number.  I encourage anyone in this discussion who wants to truly understand Side Scan Sonars and Acoustics to read the book SOUND UNDERWATER IMAGES: A guide to the Generation and Interpretation of Side Scan Sonar Data by J. P. Fish and H. A. Carr

As for the 800kHz, there is definitely a gap in the beams under the boat and therefore 800kHz DI is not showing everything below the boat.

As for the beam width of DI and the Narrow, Med, and Wide settings, this is a little more complicated.  Obviously HB has not added a third element to their units like Lowrance to achieve Down Imaging and therefore the data is derived from the SI data.  There is no way to selectively take sound returns from just a specific angle of the SI beam using a horizontal SI element as some of the diagrams posted here have depicted.  However, you can limit the range of the data being used.  I have also been discreetly told that the 2d data is being used in the processing to help filter the data so that only the SI returns that match up with the 2d returns are shown.  The 2d data is not what is being shown in DI, it is only used to help process! 

I want to reiterate, what follows is my opinion based on educated theories and certainly is not a fact, but since I know what is being recorded in the HB format and how it works, I think I have a better opinion than most here!

In Wide Mode, the DI is basically an average of both Left and Right SI elements.  That is why you will see more fish and objects suspended in the water column that may not show up on the 2d sonar.  This ultimately provides the best resolution in the water column since more data is being processed. The amount of SI data being used may be up to 2.5 times the depth of water.  So in 20' we only process the first 50' of SI returns.  The rest is not needed in DI since we don't really care whats out past 20' since we are looking for detail in the water column.  You also have to remember, that in SI what is being displayed is what we call the Slant Range data.  But that's another discussion all together.

In Medium Mode, the water column becomes filtered and we see less objects that show up outside of the 2d cone.  Now, there are a couple reasons for this.  For one, we can start to cut down the SI data by eliminating data that is considerably past our 2d range.  For example, we are in 20' of water and only want to process the SI data that extends to 1.5 times our depth.  So we only process the Si data out to 30' slant range.  HB may use a different number of course.  Now, if you notice, the bottom also starts to clear up a little and get some definition.  This is because less of the data from further out is being averaged during processing and results in a clearer image.  Now we also start to loose some of the fish or objects in the water column that are outside this range and not displayed in 2d.  Yet, most if not all of our returns in 2d are still shown in the DI.

As for Narrow Mode, the same process above is applied, but this time we only use SI data from say 1.25 times our depth, so out to 25' slant range.  We now start to see a lot more detail in the bottom and the water column has lost much of its detail except for the SI returns that match up to the 2d sonar data. 

As for my thoughts on Down Imaging, I actually use it very little in searching for shipwrecks or SAR. The Side Imaging provides me with much more detail about the objects I am looking for since they are almost always on the bottom.  I can tell more about an object from its shadow than the object itself.  I think this is also why people are flocking to DI, because it is a simpler more traditional view.  Even with my personal opinion as stated, I think DI is an excellent resource for fishermen.  It allows the user to see suspended objects in the water column in much greater detail that the traditional 2d sonar does not. And this of course is where the so called "Bread and Butter" is in this industry. So, HB DI is a big hit!

I hope this makes some sense.  I will try to come up with some fancy diagrams like Mike and Fishton to show what I am saying.  I look forward to some responses, but will not get into a debate over the methods or theory.

I must admit that by Lowrance having a third element dedicated to Down Imaging will result in better DI images.  But I think they have to have the element because their SI does not cover the full 180 degrees!

I know some HB users may not like to hear that, but I feel its true.  Don't get me wrong, I am HB through and through and in no way would switch to a Lowrance SS1.  HB has the best overall product, support, and technology.  The Side imaging with HB is unmatched. 

There are commercial units costing tens of thousand that do not compare to the HB SI units.

Sincerely,

Robert Gecy
SideImagingSoft.com Forum Administrator


SideImagingSoft.com - Humminbird Side Imaging Forums



Modified by rgecy at 1:41 AM 2/11/2010

[Kimi]

Ok. Thanks eweryone for the information.

I think its natural to sitting here now and eating up my fingernails. Years of learning and hundreds of hours and learning, and now I have to start over from a clean table. Makes a big man crying and eating sand!!!

Or maybe not.

But anyway it effects the way I am going to use my sidescan in the future. The main thing is that the systems work, and that I can get my job done. And the better I know how its working, the better.

Kimi...

[sonar2000]

Kimi.  I share your feelings.  The past two weeks have changed my thought and understanding on side imageing.  All I can do now is wait for spring and better weather to get out and learn more about HB side imageing.  The postings here have been enlightening and certainly inspired some stiumlated conversation.  All aside, my confidence in HB to provide excellent bottom representation remains high and the value of the image will be that which we need to fish, search or recover. 
I for one will be relearning sideimage and now applying that new information to the usage of our unit.  In reality, this conversation has refired my interest in bottom imageing.  Chuck

[Stymie]

OK, in a nut shell, your theory on DI is the above-mentioned slant range data processed with 2d sonar to fill in the gaps on what is under the boat and where it is in the water column?

So in a sense, they ARE using three arrays of elements, just not with any one dedicated array for the Hi-Def DI.

My 998 is still on order.

[sonar2000]

Stymie.  I am thinkiing that is the way it works.  As explained to me the 455khz datais used to scan.  Exactly how that data is collated into the DI image is still unclear but I was told the 200 or the 83 is used only for the 2d screen.  so I am guessing that just the 455 array is what is presenting the data to the program for screen image.  I have cleared my brain of all previous information and will reprogram my memory and see where we go from this.  As earlier I have all the confidence in HB to produce  the data  we will have to learn and interpret the screens.  At least we have this forum to help each other.  Keep posting.     Chuck

[FuzzyGrub]

I have also been discreetly told that the 2d data is being used in the processing to help filter the data so that only the SI returns that match up with the 2d returns are shown.  The 2d data is not what is being shown in DI, it is only used to help process! 

Modified by rgecy at 1:41 AM 2/11/2010

I think I saw some or most of that discussion on BBC!  I gave up when it hit 16-17 pages.

I am interested in the comment about the 2D assist.  I have seen a couple people mention this, but haven't seen any images that suggest how it is being used.   Also, in one of the images in that thread, the DI and 2D were offset, which turned out to be caused by using a splitter cable.   The 2D sonar was not being used in the SI transducer.  It was using descrete 2D which was a few feet offset in front of it.

If HB was using the 2D, I would expect installations like the above would cause it problems.  The particular DI images were not bad, and had similiar results to others.  I am not aware of any DI parameter that could be used for adjustment, either.

The image of discussion:


BNote: The light blue and yellow boxes in the 2D are not the exact locations, and the light blue in the SI image.
 

[Drifter]

RGecy,

One of the things I noticed (see the  the 3rd and 4th pictures above is the the DI reaches out to the full SI range and that the full screen DI displays more 'history" than the 997 SI display. I'm thinking I may be able to use this to do a quick check of what I missed sorting out other issues running the boat and watching the rods.



Water depth close to 20 feet - in fairly shallow saltwater ...






SI range 97 feet ...




DI Lower range 98 feet ...




A roughed up version of overlaying both sides, rotating and adjusting the combined image ...



Any observations and / or input welcome.

Edit: Mar. 26, 2011 ~ Updated picture links from imageevent.com to limbatus.net.

[RGecy]

Drifter,

This is exactly what I had done with a few images myself.  I captured the screen shot of the SI alone and then the DI.  Basicaly I use photoshop to blend the two, which is pretty much an average of the two images, and it came out looking very much like the DI in the Wide Setting.

There is definitely filtering going on in the Med and Narrow mode.  Its just a matter of what data is being used and how, which is going to be HBs secret!   

What is interesting is how the bottom gets more defintion in Narrow Mode, yet the water colum seems to lose some definition.  And yet Wide mode seems to give the best view for structure and suspended fish, etc. in the water column!

Robert

[RGecy]

FuzzyGrub,

What mode is the DI set to in the images you posted?  If it is in Wide Mode, I figure that they are not filtering any of the SI data.  So no matter if its in the 2d view or not, its going to be shown.

Whose file is it?  I would like to get a copy of the recording for playback and mess around with it if I can and maybe analyse it a little closer!

Robert

[Stymie]

That came from RWarner at BBC if I'm not mistaken.

[FuzzyGrub]

Stymie is correct.  I asked the same thing about the recording, but it was only a snapshot.  I wanted to investigate what the DI looked like for some of the image in the lower part of the SI.  In particular the structure that gets divided right down the middle.  According to the user, it was set to "wide" and 83Khz 2D was off. 

[sonar2000]

Darn internet and searches  here is another article in which I found the following.  Every time I find more information I get more confused. 
In describing the HB transduce. 

Two blade-shaped 80°@-10db 262kHz side-/down-looking beams:

Each beam is oriented so the shallow edge of the beam starts at the surface and extends to underneath the boat.
All beams together form 180° of coverage
Depth capability of up to 150’ for clearest images; distance out to 360’ on each side. Under optimum conditions, both greater depths and distances are possible.
Alternative 455kHz operation for these beams results in slightly less total coverage of 160°, but even higher definition images.

Chuck

[Moose1am]

Just remember that sound wave and the so called Cone don't have sharp edges.  These so called cones are more like lobes of sound.  Sound propagates in all directions unless you can focus it somehow.

So how does the Humminbird SI transducer focus the sound waves into 86 deg cones?  And How sharp are the edges of those sound waves.

Don't they use statistics to figure the size of the cones at different decibels or frequencies?


One very important thing to remember about sound waves and sonar.   Sonar can time the time it takes for a sound to go out and back to the transducer and determine it's range but it's not very good at determine the direction of the sound waves and therefore what side to put an object that's on the bottom right under the boat.  Should it put that object one the left side or the right side of the display screen?  It can't really tell which side of the boat the object was on.  It only knows the range to the object.   

 
Also remember that the new SI transducers are also looking straight down at the bottom of the lake if the transducer is mounted correctly.  Not only is the transducer sending out SI sound waves to the left and right but it's also sending the sound waves straight downward too.
 
How else could you take a recording and then later play it back using every possible view on the Humminbird SI unit? 
 
Where do you think that the unit gets it's 2 D dept data for the regular sonar screen view? 
 
Doesn't the transducer send out all it's sound waves all the time no matter what the view on the screen is set to?   
 
From what I can tell in the manual and using the 898C SI unit there is only one way to turn the unit's sonar off.  There is one setting in the menu that does this. Well other than to cut the power to the unit or turn it off with the power button.   But I didn't find any menu item that allows me to just turn off one of the crystals in the Transducer.
 
There has to be a crystal pointing sound waves straight down or they could not get the accurate 2D depth data for all the regular 2D Sonar screens. 
 
And the CPU allows us to control how much of the down imaging data is used to view the sound returns coming back and showing on the screen.  Why else would we be able to make the DI image narrow, medium or wide view?  There has to be either a filtering of the data or a changing of the sound wave lobes

[Doug Johnston]

After read this reply I believe you are very knowledgeable on the subject. I am new to the forum and have only had my HB 998c SI for a couple months. To get the most out of this, new to me, technology I would like to understand it better. I have read a lot of articles, etc. However in the std SI view I am still confused about what I am seeing in the black hole in the middle. I know it is surface to bottom, but when I see something between (bait ball, etc) what am I really looking at. Any help would be appreciated - thanks in advance.   

[RGecy]

The dark area represents the water column from surface to first sonar return.  It may not always be the bottom.  If you see something in the water column, it does not automatically mean that it is at the depth it appears on the SI view.  If your depth is say 30' and you have some fish 20' deep and then you have some fish out to the side about 20' away from the transdcuer, but only a few feet deep, they will both appear at the same depth on the SI view.  The clue to how close to the bottom the fish are would be from the shadow.  The closer an object is to its shaddow, the closer it is to the bottom.

Hope that makes sense.

Robert

[Humminbird_Greg]

As shown in the first graphic to this post: fish #2 and 3 are the same distance away from the transducer but fish #3 is farther from the bottom so its shadow will be shown much farther away from it than fish #2 which is very close to the bottom.

[felvic]

In others words, if I have a side record ( say 30m deep and 90m wide ), find a structure and want to mark a waypoint:

- If the structure is near under the boat the point is good
- If the structure is about 80m the point is good
- If the structure is between 35 and 60m the point is fake.

Am I right?

sorry for the meters, I'm french...

[RGecy]

No, thats not accurate. 

The best way to look at it, If an object is on or near the bottom, the waypoint will be accurate.  If an object is on or near the surface, it will not be accurate.  If it is directly under the boat, the waypoint will be accurate.  But just because an object shows up in the water column in SI, does not mean its under the boat.  It could be out to the side, just not as far out to the side as the water is deep.

Of course in shallow water, the difference is negligable.  In deeper water, it becomes more of an issue.

Robert

[cactusflyr]

Read through this.  Would love to have someone dissect this si pic.  Four fish...one pair on the right with shadows far away, one single on the right with close shadow (great shadow....can see fish outline), and one on the left by big rock with close shadow (almost on bottom). What are their distances from the transducer, since that's what it measures?

[FishinLite]

Here is a rough sketch of what the two fish on the right with far away shadow look like.



Based on the shadow, the two fish are probably 3 ft to the right of the boat and approx 5 ft down.