Whispers in the Walls: What the Detection Devices Actually Find, and the One Claim the Research Literature Flatly Denies

1. Introduction: the thing is inside the wall

Almost every hard problem in structural pest control has the same shape. The occupant reports something, the technician can see a surface, and the organism is behind it.

A market of instruments exists to solve that, and what they can do is consistently overstated by the people selling them. The research literature is a good deal more candid than the marketing, and this article is mostly a comparison of the two.

1.1 Why this matters here

Manitoba has almost no termites, which this journal has covered. It has a great deal of carpenter ant activity in walls, wood decay in concealed framing, and rodents in voids, and the detection problem is identical.

Almost all the published work is on termites, so the evidence has to be read across, and §22 records how far that stretches it.

2. Why visual inspection is not enough

The starting point, stated by researchers rather than by vendors.

The inspector looks for evidence of infestation such as shelter tubes, feces, damaged woods, and broken-off wings. However this method has several drawbacks, including inaccessibility of most of the buildings; uncertain level of training and motivation of the inspector; and inspection subjectivity.⁵

These problems have promoted the search for other detection methods that can detect termites in areas inaccessible to direct visual inspection and remove some of the subjectiveness.⁵

2.1 Inaccessibility of most of the buildings

Most, not some.⁵ The majority of the volume of a building is not visible: wall cavities, floor systems, roof spaces above insulation, and everything behind a finished surface.

An inspection is therefore an examination of the minority of a structure, and this journal's detection probability article set out what that does to the meaning of a negative finding.

2.2 Three separate problems

Access, competence, and subjectivity. Only the first is about the building, and the instruments being sold are usually pitched at that one.

Which is reasonable, since access is the one an instrument could plausibly fix. It is worth noticing that the other two are about the person holding it.

3. The candid part of that list

Uncertain level of training and motivation of the inspector.⁵

That is an unusually blunt sentence to find in a research paper about a commercial activity, and it names the variable this journal's articles on detection dogs and on applicator certification both arrived at from different directions.

3.1 What an instrument does and does not fix

An instrument can partly address access, by reaching where a person cannot look. It does not address motivation at all, because somebody still has to carry it, point it at the right place, and report what it said.

Section 16 shows that it may not address subjectivity either, since interpreting the output is its own judgement. That is our argument.

4. What is available

The methods, as listed in one review.

Methods developed include trained dogs, microwave, acoustic emission device, infrared temperature, moisture content, electronic odour detectors, and borescope camera.⁵

4.1 This journal has covered one already

Trained dogs, where field testing found a mean detection rate of 44 per cent against a laboratory figure of 98 per cent, and every handler believing they were above 95.

That article's finding is the right frame for this one: a detection technology's laboratory performance and its field performance are different numbers, and the gap is where the whole subject lives.

5. The thermal imaging claim

What the trade material says, quoted at length because §6 contradicts it directly.

One pest control company describes the power of infrared thermal imaging cameras, which detect the heat patterns caused by termite infestations, since termites generate heat and moisture as they build their nests and travel through wood, creating a thermal signature that these cameras can capture. Infrared imaging has revolutionized termite detection.⁷

Another states that cameras spot the subtle heat patterns associated with termite activity inside walls, and that termites generate heat and moisture as they create their tunnels, which thermal imaging can detect as warmer spots behind surfaces like wood, drywall, or paneling.⁸

5.1 The specific mechanism claimed

Both accounts assert that the insects generate heat and that the camera detects it as a warm spot.⁷⁸

That is a testable physical claim rather than marketing language, which is what makes §6 decisive.

6. What the research literature says

The contradiction, in full.

Infrared thermography is also of very limited practical utility for detecting live insects in solid wood. Since insect larvae, as ectotherms, usually do not generate relevant amounts of metabolic heat, the thermal excess relative to the surrounding wood and ambient environment is generally too small to be resolved reliably using thermographic imaging under realistic conditions. This limitation is compounded by wood's low thermal conductivity and varying moisture contents, which further complicate measurements and result in IR thermography generally being unsuited to detecting active larvae in wood.³

6.1 The argument has three parts

Ectotherms do not produce much metabolic heat, so there is little signal. Wood conducts heat badly, so what there is does not reach the surface. And wood moisture varies, so the surface temperature is noisy for reasons unrelated to insects.

Each of those alone would be a problem. Together they are the basis for the word unsuited.

6.2 The wood moisture point deserves its own note

Varying moisture content complicates the measurement.³ Wet wood and dry wood have different thermal properties, so a wall with a moisture gradient across it produces a thermal pattern with no insect involved at all.

Which means a camera pointed at a building with any moisture variation is already showing features, and the operator has to decide which ones mean something.

6.3 Which claim is wrong

The trade claim that the insects generate detectable heat.⁷⁸ An insect the size of a termite, at ambient temperature, inside a poor conductor, is not producing a warm spot on a wall.

7. Reconciling the two

The interesting part, because thermal imaging is not useless and the reason is different from the one given.

A patent specification explains: when a subterranean termite invades a structure, it brings in a substantial amount of moisture to the infested area. As the water evaporates, the infested area becomes cooler, and the difference in temperature can easily be detected by the thermal imaging camera, thus identifying a suspicious area.¹

7.1 Cooler, not warmer

The signal is evaporative cooling from water the insect carried in, and it runs in the opposite direction to the one the trade material describes.¹⁸

Which means an operator taught to look for warm spots is looking for the wrong sign of the right phenomenon.

7.2 And it only applies to some species

Subterranean termites bring moisture. Drywood termites, wood-boring beetle larvae and carpenter ants do not, and §9 makes the same point about moisture meters.

So the method is species-specific in a way that nothing in the trade description indicates.

7.3 Why the mistake is easy to make

Thermal cameras genuinely do find things in walls. They find missing insulation, air leakage, plumbing leaks and wet framing, and they are excellent at it.

An operator who has watched a camera reveal a hidden problem will generalise to hidden insects without noticing that the physics changed. The instrument works; the inference does not.

7.4 The role it can play

The patent uses the thermal scan to quickly locate potential areas before installing an acoustic sensor.¹ A screening step that narrows where to look, rather than a diagnosis.

8. The active thermography variant

A different use of the same camera.

The same specification describes detecting drywood termites concealed in a structure, involving use of a heat source to warm up the wooden structure of interest and then using a thermal imaging camera.¹

8.1 Why heating first changes the problem

Instead of waiting for a signal the insect produces, you supply energy and watch how the material releases it. A void, a gallery or a region of different density heats and cools differently from solid wood.

That is a measurement of the material rather than of the organism, which sidesteps the entire objection in §6. Our reading.

9. The moisture meter

The simplest instrument, and one whose limits are stated precisely.

The moisture meter has proven ability to detect moisture brought into hidden areas by subterranean termites but equally will not be able to detect drywood termite activity.²

9.1 It detects a consequence

Not the insect. Elevated moisture in a wall is a finding about the wall, and the inference to an organism depends on knowing what else could put water there.

This journal's articles on carpenter ants and on wood decay fungi both argued that moisture is the primary diagnostic in this climate, and the instrument that measures it is the one whose output means something here regardless of termites.

9.2 The instrument is cheap and old

A moisture meter costs a fraction of anything else in this article, requires no interpretation beyond a number, and has been in use in the building trades for decades.

It is the least glamorous item on the list, and §9.3 argues it is the most useful one here.

9.3 Why that makes it the most useful device on the list for Manitoba

Because the moisture is the actual problem. Carpenter ants excavate wood softened by moisture, decay fungi require a moisture threshold, and both are found by locating the water.

An instrument that answers the question you should be asking is better than one that answers the question you thought you were asking, and that framing is ours.

10. Acoustic detection is nearly a century old

A history that surprised us.

The use of acoustic methods for detecting insect activity in wood has a history spanning nearly a century. Researchers first reported amplified detection of movement and communication signals from termites in 1929 using repurposed parts of a telephone transmitter, and an early commercial listening device for one beetle species was marketed in Germany in the early 1950s.³

10.1 A telephone transmitter in 1929

Somebody took apart a telephone and used its microphone to listen to termites, and published.³ That is a very good piece of improvisation and it is the origin of every instrument in §4.

10.2 A commercial insect listening device in the 1950s

Which means this is not an emerging technology. It is a mature one that has been not quite working for seventy years, and §12 is the reason.

11. What the insect is producing

The signal, and its structure.

Termites make noise through their foraging activities and communication, head banging.² Stress waves caused by termites feeding are the highest amplitude stress waves produced by an active colony; thus greater sensitivity is required to detect other forms of termite activity such as their movement within an infested object.⁶

11.1 Head banging is a signal, not a noise

It is communication.² Which means part of what an acoustic detector listens for is the colony talking to itself, and that behaviour has its own rhythm rather than being continuous.

11.2 Feeding is the loudest thing they do

And it is intermittent. An insect that is not currently chewing is producing the quieter signal that requires more sensitivity to detect.⁶

So a negative reading may mean absence, or may mean the colony was resting, and this journal's detection probability article set out why that distinction has to be part of the result.

12. The sensitivity finding

The number that matters, from independent testing.

Independent studies testing the sensitivity of the instruments found that the devices, placed 50 cm from a feeding termite colony, only detected the activity about half the time. And even in those tests where the detector was placed in close proximity to the activity, the recorded count rate was only between five and twenty five counts per minute.⁶

12.1 Half the time, from fifty centimetres

Fifty centimetres is close. A wall cavity, a joist bay or a sill plate is within that distance of an accessible surface in most of the places anybody would look.

A coin flip at that range is not a diagnostic instrument, and it is remarkably close to the 44 per cent field figure this journal reported for canine detection.

12.2 What half the time does to a clearance

A single negative reading from an instrument with that detection rate carries almost no information. Two independent negatives leave a quarter of the probability mass unaccounted for, which is still a great deal.

This journal's detection probability article made exactly this argument about declaring a site free of a pest, and the arithmetic is the same whether the instrument is a dog, a trap or a sensor.

12.3 The source and its interest

This comes from the background section of a patent, which is written to establish that existing devices are inadequate and that the invention is needed.⁶

So the figure is reported by a party with a reason to report it. It is attributed to independent studies, and we would treat it as indicative rather than definitive.

12.4 Why the count rate matters too

Five to twenty-five counts per minute is a sparse signal.⁶ At the low end, that is one event every twelve seconds, which requires a long listening period to distinguish from nothing.

A technician standing in a crawlspace with a device for thirty seconds is sampling a process that may produce two or three events in that window.

13. The background noise problem

The central technical obstacle, stated in several places.

The challenge is not only detecting the sounds of termites but separating those sounds from the background noise. It's not easy; accuracy depends on the sensor, substrate characteristics and termite species, amongst a number of factors.²

Traditional approaches often suffer from low signal-to-noise ratios and environmental interference.⁹ And background events remain a concern even though the background is reduced in the ultrasonic frequency range.⁶

13.1 Substrate characteristics

The same insect in oak, in pine and in engineered wood produces a different signal at the sensor, because the material carries the vibration differently.²

So a device calibrated in one material is not calibrated in another, which is the same problem this journal's residual efficacy article found for surfaces and its automated identification article found for trap colours.

13.2 Why ultrasonic frequencies were tried

Because ordinary building noise is concentrated at lower frequencies, so listening above the audible range reduces the background.⁶

It helps and does not solve it: background events remain a concern even there, which tells you the interference is not only traffic and plumbing but the material itself responding to temperature, load and air movement.

13.3 Species characteristics

Accuracy also depends on the species.² Which means you need to know what you are listening for before you can hear it, and identifying it is what you were trying to do.

14. Where the research is going

The current direction, which is machine learning applied to the signal.

Work has covered a low-noise wireless acoustic sensing framework using adaptive noise filtering in wireless sensor networks, discriminant analysis classifying acoustic signals from one termite species emphasising the utility of frequency domain features in isolating termite activity from ambient noise, and automated frameworks based on time-frequency analysis highlighting spectral features for recognizing patterns of insect-generated sounds.⁹

14.1 Why this is a reasonable direction

The signal exists; the problem is telling it apart from everything else. That is a classification problem rather than a sensing one, and classification is where machine learning has produced its genuine successes.

14.2 The pattern is familiar

A sensing problem with a poor signal-to-noise ratio, addressed by better classification of the signal rather than by a better sensor.

This journal's article on automated insect identification found the same approach reaching 92.7 per cent precision on nine species against solid backgrounds, and degrading substantially when the background changed. We would expect the same shape of result here.

15. Microwave devices

A different physical principle, and the one with the most entertaining failure modes.

These devices detect movement behind a surface rather than sound or heat, and a university extension evaluation set out how they behave in the field.⁴

15.1 Why movement is a sensible thing to detect

Because it is the one property that distinguishes a living colony from old damage. Every visual sign this journal has described, galleries, frass, mud tubes, persists long after the insects have gone.

An instrument that reports movement is answering is it still active, which is the question that determines whether anybody needs to do anything. That is the genuine appeal and it is why §16 is disappointing.

15.2 The mounting finding

Field studies revealed that hand-held uses produce less accurate results than tripod or flap supported uses due to user hand shaking.⁴

A movement detector held in a hand detects the hand.

15.3 The angle finding

The device's output showed more noise from the user's body movement when used at 45 degrees to the inspection surface as compared to flush against the inspection surface.⁴

16. What else they detect

The false positive sources, listed by the evaluators.

Users need to stand still when reading the output or the device will pick up their body movement and produce a false positive signal. This is especially true at high sensitivities. Users also need to ensure that there are no moving objects, vehicles, plants swaying with the wind, airborne debris such as leaves and dusts, children, or animals such as pets and birds, on the other side of the inspection surface, a wall for instance, which may create false positive signals.⁴

16.1 The list is worth reading twice

Vehicles, plants swaying in the wind, airborne leaves and dust, children, pets and birds.⁴ That is not an edge case list; it is a description of an ordinary residential property on an ordinary day.

16.2 A bird on the other side of the wall

That is the detail we would want every operator to know. The instrument reports movement and does not report what moved.

16.3 And this journal has met that combination before

Article on bird mites, one article ago, described nests in soffits and wall cavities. A movement detector aimed at such a wall will read positive, and it will be reading correctly about something other than what the operator is looking for.

17. Why that matters operationally

The consequence for how a reading gets used.

A positive reading with several plausible non-target explanations requires the operator to rule them out, which means the instrument has shifted the judgement rather than removed it.

17.1 Back to subjectivity

Section 2 said instruments were developed partly to remove some of the subjectiveness of visual inspection.⁵

A device whose output depends on the operator standing still, holding it flush, and knowing whether there is a cat on the other side of the wall has relocated the subjectivity rather than eliminated it. That conclusion is ours.

17.2 What a good operator does with it

Treats a positive as a reason to investigate rather than as a finding, checks the far side of the wall before reading, mounts the device rather than holding it, and records the conditions alongside the reading.

All of which is available to anybody who has read §16, and none of which is obvious from holding the device.

17.3 The sensitivity setting

False positives from body movement are especially true at high sensitivities.⁴

So the operator chooses the trade-off between missing real activity and detecting themselves, and this journal's canine detection article found exactly that threshold effect, where false positive rate correlated with detection rate.

18. The borescope

The method that actually answers the question, and what it costs.

Borescopes are commonly used in inspections, although due to the general need to drill into materials, they are typically only used in invasive inspections. Nevertheless they can be very useful even though time-consuming, as they can provide a definitive result for the area in which they are used.²

18.1 Definitive, for one hole

That phrase contains the entire trade-off. Every other method on the list gives an uncertain answer about a large volume; this one gives a certain answer about a very small one.²

Which makes the borescope the instrument you use after something else has told you where to drill, and §7.3 is the patent describing exactly that sequence.

18.2 Why drilling is not as invasive as it sounds

A small hole in a wall cavity is repairable in minutes and is routinely made for wiring, insulation inspection and moisture testing. The reluctance is largely about the conversation with the client rather than about the damage.

Which means the instrument competing with a borescope is frequently competing with a decision nobody wanted to have rather than with a real technical constraint.

18.3 The cost is a hole

Drilling into a client's wall is a commitment. It requires permission, it requires making good afterwards, and it cannot be done everywhere in the hope of finding something.

19. The methods nobody can use

Two techniques that work and are unavailable.

Because of the health hazard X-rays pose, equipment must be operated by specialists. And computed tomography offers excellent spatial resolution but remains confined to laboratory settings and is disproportionately costly for routine infestation assessment.³

19.1 What computed tomography would give you

Excellent spatial resolution.³ A three-dimensional map of the galleries inside a timber, which would settle every question this article is about.

It is confined to laboratories and disproportionately costly, which is a statement about economics rather than about physics.

19.2 The shape of that constraint

The best imaging exists and cannot be brought to a building. Every practical method in this article is a compromise made necessary by that.

19.3 The exception where cost is no object

A niche market exists for the detection of termites and other timber pest activity in high value items such as furniture and picture frames, particularly in museums where invasive inspection techniques are simply not possible.²

Where drilling is forbidden and the object is valuable enough, the economics reverse and non-invasive methods become worth their limitations. That is the one setting where this technology is clearly the right answer.

20. A real case

A documented field example, which shows what the surrounding measurements look like.

A cupboard door in the sacristy of a church showed frass and emergence hole patterns consistent with a named beetle species. The climatic conditions showed 52 per cent relative humidity at 11 degrees Celsius with a measured wood moisture of over 15 per cent by mass.³

20.1 Note what identified it

Frass and emergence hole patterns.³ Visual evidence, of the kind §2 listed, found before any instrument was applied.

The measurements characterise the conditions rather than making the diagnosis.

20.2 Eleven degrees and 52 per cent humidity

A cool, moderately humid church interior.³ Not a tropical climate, not a leaking building, and the timber still carried enough moisture to support an active infestation.

20.3 The moisture figure

Over 15 per cent by mass,³ which is above the threshold this journal's article on wood decay identified as the point at which biological activity in timber becomes possible.

The wood was wet enough to support the problem, which is the finding that would drive any remediation.

21. What we conclude

Thermal does not see insects. It may see evaporative cooling from moisture they brought in.¹³

Acoustic is a screening tool at best. Half the time at fifty centimetres.⁶

Microwave detects movement, all of it. Including yours.⁴

Moisture is the useful measurement here. Because moisture is the actual problem in this climate.

The borescope decides. About one hole.²

And an instrument does not replace an inspection. It narrows where to look.¹

Ask what a reading would rule out. If a negative result would not change what you do, the measurement was not worth taking.

22. Limitations and open questions

Almost all of this is termite literature. Manitoba's concealed insect problems are carpenter ants, wood-boring beetles and decay-associated organisms, and we located no evaluation of these instruments against any of them.²⁵

Three sources are patent filings. Used for the moisture evaporation mechanism, the active thermography variant and the sensitivity figures, each written to establish the merits of an invention.¹⁶

Two sources are pest control company marketing. Quoted in §5 specifically in order to set them against the research literature, which is the only use we make of them.⁷⁸

The 50 per cent figure is reported second hand. Attributed to independent studies in a patent background section, and we have not read those studies.⁶

It concerns first-generation devices. The same source describes them as such, and current equipment may perform differently. We found no published field evaluation of a modern device.⁶

No cost information. These instruments range from inexpensive to very expensive and we have no figures, which is material to whether any of them is worth buying.

Sections 2.1, 3.1, 4.1, 6.1, 6.2, 7.1, 8.1, 9.1, 9.2, 11.1, 12.1, 13.1, 14.1, 16.2, 17.1, 18.1 and 20.1 are our reasoning. The three-problems framing, the argument about what an instrument fixes, the comparison with canine detection, the decomposition of the thermography objection, the identification of which claim fails, the direction-of-signal point, the active thermography reading, the consequence argument for moisture meters, the Manitoba relevance, the intermittency point, the fifty centimetre assessment, the substrate comparison, the machine learning expectation, the bird mite connection, the relocated subjectivity conclusion, the trade-off framing and the reading of the case study are ours rather than sourced positions.

23. Conclusion

Pest control companies advertise thermal cameras that detect the heat termites generate inside a wall.⁷⁸ The research literature states that larvae are ectotherms producing too little metabolic heat to resolve, that wood conducts badly and varies in moisture, and that infrared thermography is generally unsuited to detecting active larvae in wood.³ Where a camera does find something, the mechanism appears to be evaporative cooling from water a subterranean termite carried in, which makes the signal cooler rather than warmer and applies only to species that bring moisture.¹

Acoustic detection has existed since somebody wired up a telephone transmitter in 1929, and a commercial listening device was on sale in the 1950s.³ Independent testing of early instruments found them catching a feeding colony about half the time from fifty centimetres, at five to twenty-five events a minute.⁶ Microwave devices detect movement, and will report your own hand, your own body, a vehicle outside and a bird on the far side of the wall.⁴

The instruments were developed partly to remove the subjectivity of visual inspection.⁵ What they have mostly done is move it, from deciding whether there is damage to deciding what a reading means. The one device that gives a definitive answer does it by drilling a hole, and it only answers about the hole. Everything else narrows the search, which is worth having and is not the same as finding something.

References

1. Termite acoustic detection. United States patent specification. Written to establish the merits of an invention, cited as attributed material. Used for the description of a system combining a thermal scan of a structure to locate potential infestation sites with acoustic detection of termite activity sounds at those sites and comparison against a library of prerecorded sounds; for the explanation that when a subterranean termite invades a structure it brings in a substantial amount of moisture to the infested area, so that as the water evaporates the infested area becomes cooler and the difference in temperature can be detected by a thermal imaging camera, identifying a suspicious area of possible activity; for the description of an acoustic sensor in the form of an accelerometer or comparable sensor with a bandwidth of at least 100 hertz to 15 kilohertz to detect noises made by subterranean termites, with information transmitted to a portable computer and to a central operations centre; and for the separate method of detecting drywood termites concealed in a structure by using a heat source to warm the wooden structure of interest before using a thermal imaging camera. https://patents.google.com/patent/US7271706B2/en
2. Termite Detection. Professional pest management trade publication. Trade source, cited as attributed material. Used for the statements that the moisture meter has proven ability to detect moisture brought into hidden areas by subterranean termites but will not detect drywood termite activity; that borescopes are commonly used in inspections although, due to the general need to drill into materials, they are typically only used in invasive inspections, and that they can be very useful though time-consuming as they provide a definitive result for the area in which they are used; that acoustic detection is an area of active research, with termites making noise through foraging activities and communication by head banging, and research focused on both detection and monitoring; that the challenge is not only detecting the sounds of termites but separating those sounds from background noise, which is not easy, with accuracy depending on the sensor, substrate characteristics and termite species among other factors; and that a niche market exists for detection of termites and other timber pest activity in high value items such as furniture and picture frames, particularly in museums where invasive inspection techniques are simply not possible. https://professionalpestmanager.com/termite-and-pest-inspections/termite-detection/
3. Detection of Active Wood-Boring Insect Larvae Using Acoustic Emission Measurements: Principles, Experimental Validation, and Practical Applications. PubMed Central PMC13469404. Principal research source. Used for the statements that because of the health hazard X-rays pose equipment must be operated by specialists, and that computed tomography offers excellent spatial resolution but remains confined to laboratory settings and is disproportionately costly for routine infestation assessment; for the finding that infrared thermography is of very limited practical utility for detecting live insects in solid wood, since insect larvae as ectotherms usually do not generate relevant amounts of metabolic heat so that the thermal excess relative to surrounding wood and ambient environment is generally too small to be resolved reliably under realistic conditions, with this limitation compounded by wood's low thermal conductivity and varying moisture contents, resulting in thermography generally being unsuited to detecting active larvae in wood; for the history that acoustic methods for detecting insect activity in wood span nearly a century, with amplified detection of movement and communication signals from termites first reported in 1929 using repurposed parts of a telephone transmitter, and an early commercial listening device for one beetle species marketed in Germany in the early 1950s; and for the field case of a cupboard door in a church sacristy showing frass and emergence hole patterns consistent with a named beetle species, with climatic conditions of 52 per cent relative humidity at 11 degrees Celsius and measured wood moisture of over 15 per cent by mass. https://pmc.ncbi.nlm.nih.gov/articles/PMC13469404/
4. Detecting Drywood Termites in Structures with Microwave Technology. University cooperative extension urban pest programme. Used for the field study findings that hand-held use produces less accurate results than tripod or flap supported use due to user hand shaking; that the device's output showed more noise from the user's body movement when used at 45 degrees to the inspection surface compared with flush against it, noise being defined as a detected signal in the output not coming from the target insects; that users need to stand still when reading the output or the device will pick up their body movement and produce a false positive signal, especially at high sensitivities; and that users need to ensure there are no moving objects such as vehicles, plants swaying in the wind, airborne debris including leaves and dust, children, or animals such as pets and birds on the other side of the inspection surface, which may create false positive signals. https://ucanr.edu/blog/pests-urban-landscape/article/detecting-drywood-termites-structures-microwave-technology
5. Acoustic and temperature signals generated by subterranean termite infestation: its characteristics and implementations. Research publication. Used for the account that the inspector looks for evidence of infestation such as shelter tubes, faeces, damaged wood and broken-off wings, and that this method has several drawbacks including inaccessibility of most buildings, uncertain level of training and motivation of the inspector, and inspection subjectivity; for the statement that these problems have promoted the search for other detection methods able to detect termites in areas inaccessible to direct visual inspection and to remove some of the subjectiveness; and for the list of other methods developed, comprising trained dogs, microwave, acoustic emission device, infrared temperature, moisture content, electronic odour detectors and borescope camera. https://www.tandfonline.com/doi/full/10.1080/26895293.2023.2167866
6. Detection of movement of termites in wood by acoustic emission techniques. United States patent specification, background section reviewing existing instruments. Written to establish that existing devices are inadequate, cited as attributed material. Used for the statements that wood-boring and wood-feeding insects damage and destroy property and natural resources with annual direct cost measured in billions of dollars; that if wood-destroying insects are detected, various extermination methods are demonstrably effective, so that timely insect detection itself has emerged as the better part of the challenge; that stress waves caused by termites feeding are the highest amplitude stress waves produced by an active colony, so greater sensitivity is required to detect other forms of activity such as movement within an infested object; that independent studies testing the sensitivity of first-generation instruments found devices placed 50 centimetres from a feeding termite colony detected the activity only about half the time; that even where the detector was placed in close proximity to activity the recorded count rate was only between five and twenty-five counts per minute; and that background events remain a concern even though background is reduced in the ultrasonic frequency range. https://image-ppubs.uspto.gov/dirsearch-public/print/downloadPdf/6883375
7. The Science of Termite Detection Using Technology. Pest control company blog. Commercial source quoted specifically so that its claims can be set against the research literature. Used for the assertions that infrared thermal imaging cameras allow one to see through walls and uncover the stealthy movements of termites; that these devices detect the heat patterns caused by termite infestations which often go unnoticed by the naked eye; that termites generate heat and moisture as they build their nests and travel through wood, creating a thermal signature that cameras can capture; that infrared imaging has revolutionised termite detection by providing a non-invasive method to inspect homes and buildings; and that sensitive microphones designed to pick up faint noises serve as an acoustic eavesdropping tool. https://www.parkwaypestservices.com/blog/2024/april/the-science-of-termite-detection-using-technolog/
8. Protect Your Home With Advanced Termite Detection Technology. Pest control company page. Commercial source quoted specifically so that its claims can be set against the research literature. Used for the assertions that thermal imaging cameras spot the subtle heat patterns associated with termite activity inside walls or other hidden structures; that termites generate heat and moisture as they create their tunnels, which thermal imaging can detect as warmer spots behind surfaces such as wood, drywall or panelling; that this capability allows non-invasive inspections and can highlight termite problems before the infestation becomes visible; and that feeding and movement within wood create faint sounds inaudible to the human ear which can be detected by specially designed acoustic sensors using susceptible microphones tuned to the specific frequency of sounds produced. https://mintpest.com/protect-your-home-with-advanced-termite-detection-technology/
9. Early Detection of Furniture-Infesting Wood-Boring Beetles Using CNN-LSTM Networks and MFCC-Based Acoustic Features. Preprint. Used for the literature review statements that detection of termite infestations using acoustic sensing has attracted attention due to its non-invasive nature and potential for early intervention, but that traditional approaches often suffer from low signal-to-noise ratios and environmental interference, necessitating more robust detection systems; for the summary of work on a low-noise wireless acoustic sensing framework demonstrating that signal integrity could be preserved through adaptive noise filtering in wireless sensor networks; for work employing discriminant analysis to classify acoustic signals from a named termite species, emphasising the utility of frequency domain features in isolating termite activity from ambient noise; and for work proposing automated acoustic insect detection based on time-frequency analysis, highlighting the efficacy of short-term spectral features in recognising patterns of insect-generated sounds. https://arxiv.org/pdf/2507.12793